Abstract
Recently, the use of a cancer deoxyribonucleic acid (DNA) vaccine encoding tumor-associated antigens has emerged as an immunotherapeutic strategy. In this study, we monitored tumor growth inhibition by pcDNA3-hMUC1 immunization in mice using optical imaging. To determine the anti-hMUC1-associated immune response generated by pcDNA3.1 or pcDNA3-hMUC1, we determined the concentration of interferon-γ (IFN-γ) protein and CD8+IFN-γ cell numbers among lymphocytes from the draining lymph nodes of mice immunized with pcDNA3.1 or pcDNA3-hMUC1. After subcutaneously injecting CT26/hMUC1-Fluc into mice immunized with pcDNA3-hMUC1, we monitored in vivo tumor growth inhibition using an optical imaging method. The concentration of IFN-γ protein in pcDNA3-hMUC1 was higher than that of the pcDNA3.1 group (2.7 ⩽ 0.08 ng/mL and 1.6 ± 0.07 ng/mL, respectively, p < .001. The number of hMUC1-associated CD8+IFN-γ cells in pcDNA3-hMUC1-immunized animals was 30-fold higher than in the pcDNA3.1 group. Bioluminescent images showed tumor growth inhibition in pcDNA3-hMUC1 immunized animals up to 25 days after immunization. A good correlation (r2 = .9076: pcDNA3/hMUC1 group; r2 = .7428: pcDNA3.1 group) was observed between bioluminescence signals and tumor weights in two mice in each group. We conclude that optical bioluminescent imaging offers a useful means of monitoring the antitumor effects of cancer DNA immunization in living animals.
One example is cancer vaccine for human MUC1 mucin, which is overexpressed in an incompletely glycosylated form in various human cancers. This underglycosylated form exposes its immunogenic 20-amino acid repeat sequences and evokes immune responses against cancer cells expressing it.6,7 Because the high expression of the tumor-associated antigen hMUC1 is related to rapid tumor progression and a corresponding poor prognosis in several types of human cancer, it is considered an attractive immunotherapeutic target. Cancer DNA vaccine encoding hMUC1 has been studied for targeting epithelial cancers expressing high levels of MUC1.8,9
However, cancer vaccine researchers have encountered problems concerning the evaluation of the therapeutic effects of such therapies. The measuring of tumor dimensions using a caliper is limited in terms of eliminating nonviable tumor tissues, and weighing tumors and determining real sizes can be performed accurately only by excision postmortem. In addition, even this method cannot accurately determine actual tumor burdens owing to reactive non-neoplastic cell infiltration, necrosis, and fibrosis within tumor masses, and evaluations of therapeutic effects are difficult in vivo in animal experiments.
Thus, a new method of observing tumor growth using noninvasive methods is required in living animals. Some groups have developed noninvasive in vivo imaging methods, so-called reporter gene imaging, and have successfully monitored tumor growth in living animals.10–13 We also developed a dual imaging method based on the use of the luciferase gene and found that this imaging method reflects viable cancer cell numbers and that it can be used to evaluate cell number changes after doxorubicin treatment. 14
Given that in vivo reporter gene imaging represents the exact size of a tumor mass, we can examine the therapeutic effects of cancer DNA vaccine immunization at earlier time points without sacrificing experimental animals. In this study, we monitored noninvasively and repetitively the therapeutic effects of pcDNA3-hMUC1 vaccine on hMUC1- and Fluc-expressing tumors in living animals using the devised bioluminescent optical imaging technique.
Materials and Methods
Animals
Specific pathogen-free 6-week-old female BALB/c mice were obtained from SLC Inc. (Hamamatsu, Japan). All experimental animals were housed under specific pathogen-free conditions and handled in accordance with the guidelines of the Seoul National University Animal Research Committee.
Generation of Complementary DNA Constructs and Plasmid Preparation
The human pancreatic mucin1 gene hMUC1 (accession no. J05582) was cloned into the BamHI site of the pcDNA3 vector (Invitrogen, Carlsbad, CA). 15 Plasmid DNA was amplified in Escherichia coli DH5α and purified by large-scale plasmid preparations using endotoxin-free Giga Prep columns (Qiagen, Chatsworth, CA). DNA was dissolved in endotoxin-free buffer for storage.
Generation of Retrovirus
The firefly luciferase gene under the EF-1α promoter was cloned into pMSCVneo (BD Bioscience Clontech, Palo, Alto, CA). Retrovirus was produced by cotransfection into a human 293FT producer cell line using a retroviral vector carrying the luciferase gene and packaging plasmids (gag, pol, vsv-g) (BD Bioscience Clontech). Transient transfection using 36 μl of Lipofectamine 2000 (Invitrogen, Carlsbad, CA) was performed using 6 × 106 293FT cells in 10 cm tissue culture plates. Cells were co-transfected with 5 μg of retroviral vector and 10 μg of packaging plasmids. Growth media (Dulbecco's Modified Eagle's Medium containing 10% fetal bovine serum [FBS] and 1% penicillin-streptomycin) were changed at 24 hours posttransfection, and the retrovirus-containing supernatant was harvested at 48 hours post-transfection. Harvested supernatant was centrifuged at 3,000 rpm for 15 minutes at 4 °C to pellet cell debris and stored at −80°C for later use. To assess the activity of recombinant virus encoding the firefly luciferase reporter gene, HT1080 human fibrosarcoma cells were infected by adding the thawed retrovirus-containing supernatant with 10 μg/ml of Polybrene (BD Bioscience Clontech). The luciferase expression of cells was confirmed by checking luciferase activity using a luminometer (Applied Biosystem, Foster City, CA).
Murine Tumor Cell Line Expressing hMUC1 and Firefly Luciferase
hMUC1 expressing CT26 (CT26/hMUC1) was kindly provided by Dr. Jung-Ah Cho. 15 Viral supernatants were transduced into CT26/hMUC1 that has an H-2d MHC type. Stable clones (CT26/hMUC1-Fluc) were also subjected to luciferase assays using a microplate luminometer (TR717, Applied Biosystem).
Western Blot Analysis
Cells were lysed in radioimmunoprecipitation assay buffer (50 mM Tris-HCl; 150 mM NaCl; 1% NP-40; 0.25% Nadeoxycholate; 1 mM ethylenediaminetetraacetic acid; 1 mM phenylmethanesulfonylfluoride; 1 μg/mL of aprotinin, leupeptin, and pepstain; 1 mM Na3VO4; and 1 mM NaF) for 10 minutes on ice and then centrifuged at 14,000g at 4°C to remove cell debris. The supernatant was used to quantify the prepared proteins by the BCA method (Pierce, Thermo Fisher Scientific, Rockford, IL). Unless specifically mentioned, 20 μg of proteins was resolved by 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). Proteins were subsequently transferred to a membrane (Millipore, Billerica, MA) and blocked in 5% skim milk in TTBS (10 mM Tris, 150 mM NaCl, 0.1% Tween 20) for 1 hour at room temperature. The membranes were incubated with the primary antibodies for 1 hour at room temperature. Anti-MUC1 (Biomeda, Foster City, CA) antibody was used. After 1 hour, blots were vigorously washed in TTBS and then incubated with horseradish peroxidase (HRP)-tagged secondary antibody for 30 minutes. The labeled proteins were visualized by enhanced chemiluminescence using the ECL kit (Amersham Biosciences, Piscataway, NJ).
Fluorescence-Activated Cell Sorting Analysis
To examine hMUC1 expressions on the surfaces of CT26/MUC1-Fluc, cells were harvested and suspended in 0.1% bovine serum albumin-containing phosphate buffered saline (PBS). Primary antibodies were added to these suspensions; anti-MUC1 (Biomeda) mouse antibody was used as the primary antibody. 16 After 1 hour's incubation on ice, cells were washed and pelleted to remove unbound antibodies. FITC-tagged antimouse antibody was used as a secondary antibody and incubated for 30 minutes on ice. Cold 2% paraformaldehyde (PFA)-containing PBS was used to fix the cells, and, finally, fluorescence intensities were measured using a Coulter FACScan. Fluorescence-activated cell sorting (FACS) analysis showed that CT26/hMUC1-Fluc highly expressed hMUC1 tumor antigen (Figure 1A). In addition, we observed that bioluminescent signals increased according to cell numbers (Figure 1C).
Establishment of a stable cell line expressing hMUC1 and the firefly luciferase gene. CT26/hMUC1 was transduced with a retroviral construct containing the Fluc gene driven by the EF1α promoter. A, hMUC1 gene expression was confirmed by fluorescence-activated cell sorting. CT26/hMUC1-Fluc were incubated with antihuman MUC1 antibody and then with fluorescein isothiocyanate (FITC)-labeled antimouse antibodies. As a control, cells were stained with FITC-labeled secondary antibody only. B, Twenty micrograms of total cellular proteins was resolved by SDS-PAGE and subjected to Western blotting analysis using anti-MUC1 antibody. C, Fluc gene expressions were determined using in vitro luciferase assays. Bioluminescence signals correlated linearly with cell numbers. Bars represent means ± SD. RLU = relative light unit.
Cytokine Analysis
pcDNA3.1 (50 μg/100 μL) or pcDNA3-hMUC1 (50 μg/100 μL) was injected intramuscularly into the quadriceps muscles of both hind legs of mice once a week for 2 weeks. Draining lymph nodes in these mice were extracted. Lymphoid cells of draining lymph nodes were isolated in RPMI complete medium (RPMI 1640, 10% FBS). Twenty-four-well plates were layered with 105 of hMUC1 expressing CT26 and 106 of lymphoid cells at 37°C for 2 days. The plates were then centrifuged, and supernatants were transferred to 96-well plates and incubated at 4°C overnight. After washing with washing buffer (PBS supplemented with 0.5% Tween 20), wells were incubated with 3% blocking buffer (BSA in PBS containing 0.5% Tween 20) for 1 hour. Second layers consisting of either rat antimouse interleukin−10 (IL−10) (BD Pharmingen, NJ) or rat antimouse interferon-γ (IFN-γ) (BD Pharmingen) were then applied to plates for 1 hour. Plates were then washed, and the third layer consisting of rabbit-antirat immunoglobulin-HRP (Dako, Glostrup, Denmark) was added. The reaction was allowed to develop and was read using an enzyme-linked immunosorbent assay reader according to the manufacturer's instructions.
Intracellular Cytokine Staining and Flow Cytometric Analysis
pcDNA3.1 (50 μg/100 μL) or pcDNA3-hMUC1 (50 μg/100 μL) was injected intramuscularly into the quadriceps muscles of both hind legs of mice once a week for 2 weeks. Draining lymphoid cells in immunized mice were isolated from these mice and restimulated in vitro for 48 hours at 1 × 106 cells/mL with irradiated (50 Gy) CT26/hMUC1 cells (1 × 105cells/mL), and Golgistop (BD Pharmingen) was added for the last 6 hours of this restimulation. The cells were then washed once in FACS buffer and stained with FITC-conjugated monoclonal rat antimouse CD8 or CD4 antibody (BD Pharmingen). Cells were immunostained for cytokines using a Cytofix/cytoperm kit (BD Pharmingen) and were also stained with phycoerythrin-conjugated anti-IFN-γ (BD Pharmingen). Flow cytometric analysis was performed on a Becton, Dickinson FACScan using CELLQuest software (Becton, Dickinson Immunocytometry Systems, Franklin Lakes, NJ).
Monitoring of Tumor Growth Inhibition in Living Mice
The IVIS200 imaging system (Xenogen, Alameda, CA), which included an optical charge-coupled device (CCD) camera mounted on a light-tight specimen chamber, was used for data acquisition and analysis. Firefly
pcDNA3.1 (50 μg/100 μL) or pcDNA3-hMUC1 (50 μg/100 μL) was injected intramuscularly into the quadriceps muscles of both hind legs once a week for 2 weeks. One week after the final immunization, each group was inoculated subcutaneously. with 1 × 105 CT26/hMUC1-Fluc in right thighs. Mice were repeatedly imaged at 2, 7, 11, 18, and 25 days after tumor injection using an optical CCD camera to acquire photons 10 minutes, after injecting
Statistical Analysis
The Student t-test (two-tailed) was used throughout, and p values of < .05 were considered to be statistically significant.
Results
Stable Cell Line Expressing hMUC1 and Firefly Luciferase Gene
We obtained stable cell lines expressing hMUC1 gene on their cell surfaces as shown by FACS and Western blot analysis (see Figure 1, A and B). Also, it was shown by in vitro luciferase assay that bioluminescence signals were correlated with cell numbers (see Figure 1C).
hMUC1-Associated CD8+IFN-γ+ Cells in the Draining Lymph Nodes of Mice Immunized with pcDNA3-hMUC1 or pcDNA3.1
The numbers of CD8+IFN-γ+ cells in mice immunized with pcDNA3-hMUC1 were 30-fold higher than in mice immunized with pcDNA3.1 (Figure 2; p < .005). When lymphoid cells were re-stimulated with CT26, no significant difference in CD8+IFN-γ+T cell counts was observed between the pcDNA3-hMUC1 and pcDNA3.1 groups, and no difference was found between the pcDNA3.1 and pcDNA3-hMUC1 groups regarding the proportions of CD4+IFN-γ+cells among cells obtained from the draining lymph nodes of immunized mice (data not shown).
Cytokine Profiles
The concentration of IFN-γ protein, a T helper 1 (Th1) cytokine, in the pcDNA3-hMUC1 group was higher than in the pcDNA3.1 group (Figure 3A; p < .0005). However, the concentration of IL-10 protein, which is related to the T helper 2 (Th2) cytokine, was not different in the two groups (Figure 3B).
Determination of CD8+IFN-γ cell numbers in the draining lymph nodes of mice immunized with pcDNA3-hMUC1. Mice were immunized intramuscularly with pcDNA3.1 or pcDNA3-hMUC1 once a week for 2 weeks. Draining lymph nodes were removed from immunized mice at 2 weeks postvaccination and restimulated in vitro against CT26/hMUC1-Fluc for 2 days. Restimulated draining lymphoid cells were then stained using CD8 surface marker and intracellular interferon-γ (IFN-γ). The double staining results (A and B) shown here are representative of two experiments. Bars represent means ± SD; n = 7 mice/group.
Monitoring Tumor Growth Inhibition due to pcDNA3-hMUC1 Immunization by Optical Imaging
Bioluminescent images showed tumor growth inhibition in the pcDNA3-hMUC1 group but not in the pcDNA3.1 group from 11 days postinoculation (Figure 4A). Tumor growth inhibition was sustained until 25 days post-tumor challenge. The total photon efflux of the tumor region of interest in the pcDNA3/hMUC1 group was significantly lower than that in the pcDNA3.1 group (Figure 4B; p < .05). Also, we observed that the pcDNA3-hMUC1 immunization had a therapeutic effect by comparing the weighted tumor masses of the two groups (Figure 4C; p < .005). A moderate correlation (r2 = .9076: pcDNA3/hMUC1 group; r2 = .7428: pcDNA3.1 group) was observed between bioluminescence and tumor weights in both groups (Figure 4, D and E).
Discussion
In vivo molecular imaging using reporter genes was developed to monitor gene expressions, signal transduction, and cell trafficking in living organisms.17–22 Moreover, because in vivo imaging can reveal details of the localizations, intensities, and durations of the expressions of various target genes in living cells, these techniques have been used to visualize the gene expression of cancer DNA vaccine, tumor cell growth, and therapeutic effects in vivo. Several methods can be used for reporter gene imaging, that is, fluorescence, bioluminescence, magnetic resonance imaging, and nuclear medicine imaging modalities. Previously, we successfully visualized the duration and localization of DNA vaccine gene expression at not only the injection site but also the draining lymph nodes from 10 hours to 14 days by using bioluminescent images following pcDNA3.1-Fluc injection in immunocompetent BALB/c mice. 23 In addition, we transfected a firefly luciferase gene to SK-HEP1 (human hepatocellular carcinoma) cancer cells and found the very good correlation between the number of SK-HEP1 cancer cells and luciferase intensity in both in vitro and in vivo models.14,24
Cytokine analysis after the intramuscular immunization of mice with pcDNA3-hMUC1. Lymphoid cells from mice immunized with pcDNA3-hMUC1 or pcDNA3.1 were stimulated with irradiated CT26/hMUC1-Fluc for 2 days, and supernatants were then assayed for the presence of interferon-γ or interleukin-10. Experiments were performed in duplicate. Bars represent means ± SD; n = 7 mice/group.
In this study, we adopted an optical imaging using a firefly luciferase rather than other in vivo imaging modalities. This is because light emitted from luciferase is easily detected with a CCD camera; thus, it allows the visualization of diverse phenomena in living animals. The advantages of this bioluminescent system are its high sensitivity and simplicity; however, it can be used only in small animals owing to tissue attenuation.
In vivo visualization of tumor growth inhibition caused by pcDNA3-hMUC1 immunization using hMUC1 and Fluc gene expressions. A, Mice were immunized with pcDNA3.1 or pcDNA3-hMUC1 once a week for 2 weeks. One week after the final immunization, animals in both groups were inoculated subcutaneously with 1 × 105 CT26/hMUC1-Fluc in right thighs. Images were obtained at 5 minutes after tumor challenge on day 1 until day 25. The red arrow indicates the cancer cell injection site. At 2 days postchallenge, no bioluminescent signals were observed at the injection site because of scale bar adjustment. B, To quantify light intensities, regions of interest were drawn over tumor regions and total photon efflux was measured. C, At 25 days postchallenge, tumors were excised and weighed. D and E, The correlation between the bioluminescence of tumors and tumor weights was calculated. Experiments were performed in duplicate. Bars represent means ± SD; n = 7 mice/group.
In the present study, we report tumor growth inhibition caused by intramuscular pcDNA3-hMUC1 immunization into mice, as determined by optical imaging. Because stable CT26/hMUC-Fluc cells expressed both immunogenic target and reporter genes in the same cells, we were able to observe the therapeutic effects of hMUC1 vaccination in living mice. The bioluminescent signals of tumor masses increased more rapidly in the pcDNA3.1 group than in the pcDNA3-hMUC1 group from 11 days after vaccination.
Although optical imaging effectively allowed the monitoring of tumor growth patterns after tumor-associated antigen immunization, other immunologic methods may be required to support optical imaging data. The number of CD8+IFN-γ cells in the pcDNA3-hMUC1 group was 30fold higher than in the pcDNA3.1 group. Moreover, the concentration of IFN-γ protein in the supernatant of stimulated lymphoid cells from the pcDNA3-hMUC1 group was higher than in that from the pcDNA3.1 group (see Figure 3). Because several studies have reported that IFN-γ and IL-10 are related to Th1 and Th2 immune responses, respectively,25,26 we speculate that the main effect of this kind of immunotherapy is the Th1 immune response.
In summary, using the devised optical imaging method, we evaluated the therapeutic effects of pcDNA3-hMUC1 immunization on tumors expressing both the hMUC1 and Fluc genes in a mouse model. Bioluminescent signals illustrated tumor growth inhibition in mice immunized with pcDNA3-hMUC1 over a period of 25 days. In addition, we demonstrated the presence of an anti-hMUC1-associated immune response both in terms of CD8+IFN-γ cell numbers and according to secreted IFN-γ levels. We conclude that bioluminescent optical imaging may prove useful for optimizing dosages, determining administration frequencies, and evaluating the efficacies of adjuvant treatments during the development of cancer DNA vaccine protocols.