Multimodality Imaging of Bone Marrow-Derived Dendritic Cell Migration and Antitumor Immunity.

Here, we sought to monitor bone marrow-derived dendritic cell (BMDC) migration and antitumor effects using a multimodal reporter imaging strategy in living mice. BMDCs were transduced with retroviral vector harboring human sodium iodide symporter (hNIS, nuclear imaging reporter), firefly luc2 (optical imaging reporter), and thy1.1 (surrogate marker of NIS and luc2) genes (BMDC/NF cells). No significant differences in biological functions, including cell proliferation, antigen uptake, phenotype expression, and migration ability, were observed between BMDC and BMDC/NF cells. Combined bioluminescence imaging and I-124 positron emission tomography/computed tomography clearly revealed the migration of BMDC/NF cells to draining popliteal lymph nodes at day 7 postinjection. Interestingly, marked tumor protection was observed in mice immunized with TC-1 lysate-pulsed BMDC/NF cells. Our findings suggested that multimodal reporter gene imaging of NIS and luciferase could provide insights into the biological behaviors of dendritic cells in living organisms and could be a useful tool for the optimization of DC-based immunotherapy protocols.

Introduction

Bone marrow–derived dendritic cells (BMDCs) are professional antigen-presenting cells that express several specialized molecules, including major histocompatibility complex (MHC) class I/II molecules, costimulatory molecules (B7), and adhesion molecules (intracellular adhesion molecule [ICAM]-1, ICAM-3, and lymphocyte function–associated antigen 3), and that stimulate effectors such as CD4 and CD8 T cells [1], [2], [3]. Because of their unique characteristics, BMDCs can recognize various tumor antigenic peptides, including carcinoembryonic antigen and human papilloma virus 16 E6/E7, resulting in strong antitumor immunity [4], [5], [6]. Although several cancer therapeutic strategies using DC-based immunotherapy alone or in combination with radiation therapy and chemotherapy have been developed, these treatments are only partially effective because there are currently no techniques to monitor the biological behaviors of DCs in vivo. Therefore, to optimize DC-based immunotherapy strategies, it is necessary to develop reliable and quantitative imaging tools that enable monitoring the localization, distribution, and migration of infused DCs in vivo.

Molecular imaging with reporter genes to observe the proliferation, localization, and migration of cancer cells, stem cells, and immune cells has been widely investigated [7], [8], [9]. In vivo bioluminescent imaging (BLI) with firefly and Renilla luciferases has multiple advantages, such as high sensitivity, ease of use, cost-effectiveness, and no requirement for complex facilities. However, BLI is associated with limited tissue penetration depth; thus, it is difficult to detect cells of interest in deep tissues. Nuclear medicine imaging of sodium iodide symporter (NIS) and herpes simplex virus type 1 thymidine kinase (HSV1-tk) gene expression has great sensitivity and no depth limitation; however, this method cannot be used to distinguish small populations of cells located at a specific site of interest because of high background activity resulting from physiological tracer uptake by surrounding tissue. To overcome these limitations of single imaging modalities, researchers have attempted to develop multimodal reporter gene imaging strategies that can be used for tracking early distribution and migration of cells in deep tissues [10], [11], [12], [13], [14].

In this study, we attempted to visualize the migration of primary BMDCs to lymphoid organs, particularly draining lymph nodes, using a multimodal reporter gene system involving NIS and luc2 genes in living mice. Furthermore, we extensively investigated the antitumor immunity induced by vaccination of mice grafted with TC-1 cancer cells expressing Renilla luciferase with BMDCs expressing the multimodal reporter, an important concept in the clinical application of multimodal reporter systems.

Materials and Methods

Ethics Statement

All described procedures were reviewed and approved by Kyungpook National University (KNU-2012-43) Animal Care and Use Committee and performed in accordance with the Guiding Principles for the Care and Use of Laboratory Animals.

Transduction of Multimodal Reporter Genes into BMDCs

A retroviral vector harboring hNIS, firefly luc2, and Thy1.1 genes termed “Retro-NFT” was established by CosmoGenetech Co. Ltd. (Seoul, South Korea) using a pCHAC backbone (Allele Biotechnology, San Diego, CA). Each reporter gene was linked with a T2A sequence to allow effective and independent reporter expression (Figure 1A).

Figure 1

Schematic representation of the principle and multimodal reporter gene construct for in vivo tracking of DCs. The multimodal reporter system involves expression of the hNIS, luc2, and Thy1.1 genes, which act as a nuclear reporter, optical reporter, and surrogate for hNIS and luc2, respectively.

Gryphon packaging cells (Allele Biotechnology, San Diego, CA) were cultured in Dulbecco's modified Eagle's medium (Life Technologies, Grand Island, NY) with 10% FBS. Ecotropic retrovirus particles were generated by transfecting the Gryphon cells with Retro-NFT using a calcium phosphate transfection kit (Invitrogen, Carlsbad, CA). Three days later, fresh viral supernatant was used for spin infection (2000 rpm, 90 minutes, 32°C) of BMDCs, and infected cells were plated in 100-mm Petri dishes containing complete medium. BMDCs expressing Retro-NFT are hereafter referred to as “BMDC/NF cells.” After staining of the cells with an APC-Cy7–conjugated Thy1.1 antibody (BD Biosciences, San Jose, CA), Thy1.1-positive cells were determined using a BD Accuri C6 flow cytometer (BD Biosciences).

Reporter Gene Assay

For I-125 uptake assays, the indicated number of cells was plated in 24-well plates. One day after seeding, the cells were incubated with 500 μl of Hank's balanced salt solution (HBSS, HyClone, Logan, UT, USA) containing 0.5% bovine serum albumin (bHBSS), 3.7 kBq carrier-free I-125 (PerkinElmer), and 10 μM sodium iodide (specific activity 740 MBq/mmol) at 37°C for 30 minutes. After incubation, the cells were washed twice as quickly as possible with ice-cold bHBSS buffer and lysed in 500 μl of 2% sodium dodecyl sulfate. Radioactivity was measured using a Packard Cobra II gamma-counter (PerkinElmer, Waltham, MA).

For luciferase assays, cells were plated in white and clear-bottom 96-well plates. The appropriate substrate was added to each well 24 hours later, and BLI signals were measured using a microplate reader (BMG Labtech, Offenburg, Germany) or IVIS Lumina III (PerkinElmer).

Proliferation Assay

Cell proliferation was assayed using a Cell Counting Kit (CCK)-8 (Dojindo Laboratories, Tokyo, Japan). BMDCs and BMDC/NF cells were plated at 1 × 104 cells per well in 96-well plates. Ten microliters of CCK-8 solution was added to each well at 24 and 48 hours after cell seeding, after which the plates were incubated at 37°C for 3 hours. The absorbance at 450 nm was measured using a microplate reader (BMG Labtech).

Flow-Cytometric Analysis

BMDCs and BMDC/NF cells were stained with PE-conjugated anti-CD54, anti-CD86, anti–H-2Kb (MHC class I,) or anti–I-A/I-E (MHC class II); FITC-conjugated anti-CD11c; or APC-Cy7–conjugated anti-CD197 (CCR-7) antibodies (BD Biosciences) at 4°C for 30 minutes. After washing with PBS, the cells were analyzed using a BD Accuri C6 flow cytometer. Isotype-matched monoclonal antibodies were used as controls. The data were analyzed using the FlowJo analysis software (FlowJo, Ashland, OR).

Transwell Migration Assay

BMDCs and BMDC/NF cells were incubated with 1.0 μg/ml of prostaglandin E2 (PGE2; Sigma-Aldrich, St. Louis, MO) for 24 hours. Then, culture medium containing CCL19 (PeproTech, Rocky Hill, NJ) was added to the lower chamber, and 2 × 105 cells were added to the upper chamber, which contained polycarbonate membranes with 5-μm pores (BD Biosciences). After a 2-hour incubation at 37°C, the migrated cells were collected and counted by flow cytometry. Events acquired for 60 seconds were analyzed using FlowJo.

Antigen Uptake Efficiency

BMDCs and BMDC/NF cells were incubated with 100 μg/ml of Texas Red–ovalbumin (Life Technologies) at 4°C or 37°C for 1 hour. After labeling, the cells were washed with PBS and analyzed using a BD Accuri C6 flow cytometer.

In Vivo Therapy

The scheme for in vivo therapy is outlined in Figure S2. Total lysates of TC-1/Rluc cells were prepared by freezing and thawing thrice. Fifty micrograms of each lysate was incubated with 2 × 106 BMDC/NF cells overnight, and pulsed or unpulsed BMDC/NF cells were intramuscularly injected into the hind legs of mice once a week for 2 weeks (n = 7 mice/group). After the last immunization, the mice were challenged with TC-1/luciferase, and tumor progression was monitored with BLI. After imaging, the mice were sacrificed. The tumors were excised, and the tumor weights were measured.

In Vivo Imaging of DC Migration to Lymph Nodes

Mice (n = 5) received 30 ng TNF-α via subcutaneous injection into the hind footpads. One day later, 1 × 106 BMDCs and BMDC/NF cells were subcutaneously injected into the left and right footpads of the hind leg, respectively, and combined optical and positron emission tomography (PET)/computed tomography (CT) imaging was performed at the indicated times.

After imaging, the mice were sacrificed. The DPLNs and control popliteal lymph nodes were excised, placed on a black sheet, and subjected to BLI ex vivo.

Cytotoxic T Lymphocyte (CTL) Assay

Splenocytes harvested from tumor-bearing mice were restimulated with total TC-1 lysates and interleukin (IL)-2 for 4 days. Then, the cells were washed twice with PBS. TC-1/Rluc cells were used as target cells. Target cells were seeded in black, clear-bottomed 96-well plates (1 × 104 cells/well). After overnight incubation, splenocytes were added as effectors to the wells at an effector/target ratio of 5:1. Eleven hours later, luciferase activity was measured with an IVIS Lumina III instrument (PerkinElmer).

Statistical Analysis

All data are expressed as the mean ± standard deviation (SD) from at least three representative experiments, and statistical significance was determined using an unpaired Student’s t test with GraphPad Prism 5. P values < .05 were considered statistically significant.

Results and Discussion

Design of the Retro-NFT Multimodal Reporter System Harboring hNIS, luc2, and Thy1.1

For in vivo DC tracking, we designed a new multimodal reporter system, Retro-NFT, based on a retroviral vector harboring NIS as a nuclear imaging reporter, firefly luc2 as an optical reporter, and Thy1.1 as a surrogate marker for NIS and luc2 (Figure 1). For effective protein expression, each reporter gene was linked with self-cleaving 2A peptides, which allow for highly efficient cleavage.

Evaluation of the Retro-NFT Multimodal Reporter System in Gryphon E Cells

After transfection of Gryphon E cells with Retro-NFT, we could detect mRNA expression of the respective reporters by reverse transcription polymerase chain reaction analysis (data not shown). Fluorescence-activated cell sorting analysis with anti-Thy1.1 revealed Thy1.1 protein expression in 90.8% of Retro-NFT–transfected cells (Figure 2A). Furthermore, there was an increase in iodide uptake in transfected but not parental cells (Figure 2B). Luciferase assays showed a cell number–dependent increase in signal in transfected but not parental cells (Figure 2C).

Figure 2

Expression characteristics of the multimodal reporter gene system for in vivo tracking of DCs. The multimodal retroviral vector construct was transfected into virus-producing Gryphon E cells, and the expression of each protein was determined by (A) fluorescence-activated cell sorting analysis with anti-Thy1.1 antibodies (red: isotype-matched cells, blue: anti-Thy1.1-stained cells), (B) iodide uptake for NIS, and (C) luciferase assays in 293FT and 293FT/NFT cells. Data indicate means ± SDs.

Introduction of the Multimodal Reporter Genes into BMDCs

On day 3 posttransduction of BMDCs to generate BMCD/NF cells, significant Thy1.1 protein expression was detected in 18.2% of the cells (Figure 3A). Because Thy1.1 acted as a surrogate marker for hNIS and luc2, the population expressing Thy1.1 indirectly represented hNIS- and luc2-positive BMDC/NF cells. Immunoblotting with anti-NIS and anti-luc2 antibodies showed specific bands corresponding to the respective reporters in BMDC/NF cells but not in BMDCs (Figure 3B).

Figure 3

Expression of the multimodal reporter by primary BMDCs. (A) Flow-cytometric analysis of BMDC/NF cells using anti-Thy1.1 antibodies (red: isotype-matched cells, blue: anti-Thy1.1–stained cells). (B) Immunoblot analysis of hNIS and luc2 protein expression with anti-NIS or anti-luciferase antibodies, (C) iodide uptake (KClO4 as an NIS inhibitor), and (D) in vitro BLI in BMDCs and BMDC/NF cells. Experiments were performed at least in triplicate, and bar graphs represent means ± SDs.

Next, we examined hNIS and luc2 protein activity by iodide uptake assays and in vitro BLI. As shown in Figure 3C, iodide uptake was increased by five-fold in BMDC/NF cells, whereas the increase was completely inhibited by treatment with the NIS inhibitor KClO4. In vitro BLI revealed that the luciferase signal was increased in a cell number–dependent manner in BMDC/NF cells but not in BMDCs (Figure 3D).

Next, we determined whether the radioactive and optical signals from reporter BMDCs correlated with the number of injected reporter BMDCs in mice. As shown in Figure 4A, BLI and radioactive signals increased in a cell number–dependent manner following subcutaneous injection. A good correlation (r2 = 0.8618 and 0.8623 for BLI and PET imaging, respectively) was observed between the number of DCs and BLI and radioactive signals from reporter BMDCs (Figure 4, B and C).

Figure 4

Visualization of BMDC/NF cells following subcutaneous injection. (A) BLI and I-124 PET/CT imaging of different numbers of subcutaneously injected BMDCs or BMDC/NF cells. (B) Correlation between the number of injected BMDC/NF cells and BLI or radioactive signal. Bar graphs represent means ± SDs.

Effects of Multimodal Reporter Genes on Biological Function of BMDCs

Molecular imaging with reporter genes has been used to track circulating cancer cells and to determine the biological behaviors of immune and stem cells in living mice [7], [9], [15], [16], [17], [18]. For successful application of reporter genes to DC tracking in vivo, the reporter should not induce adverse effects on cell proliferation or biological functions.

To this end, we first compared proliferation between parental BMDCs and BMDC/NF cells. CCK-8 assays indicated no significant differences between the two groups (Figure 5A). Next, we investigated the effects of reporter construct introduction on antigen uptake, phenotype marker expression, and Transwell migration of DCs. The results showed no differences in uptake of the OVA peptide at 37°C between groups; however, antigen uptake at 4°C was markedly inhibited in both groups (Figure 5B). Phenotypic analysis demonstrated no significant differences in the expression of the DC-specific markers CD11c, CD54, CD86, MHC class I and II, and CCR7 between BMDCs and BMDC/NF cells (Figure 5C). In Transwell migration assays, there was a distinct increase in the migration of BMDCs in the presence of PGE2, which has been shown to upregulate the expression of CCR7 in DCs and may increase the responsiveness of DCs to C-C chemokine ligands [19], with no difference between BMDCs and BMDC/NF cells (Figure 5D).

Figure 5

Effects of multimodal reporter expression on biological functions of BMDCs. (A) Cell proliferation assay with CCK-8 reagents; (B) antigen uptake assay with Texas Red–ovalbumin; (C) phenotypic marker expression analyzed using specific antibodies; and (D) Transwell migration assay for BMDCs and BMDC/NF cells. Experiments were performed at least in triplicate, and bar graphs represent means ± SDs. N.S. = not significant.

These results indicated that the introduction of the multimodal reporter genes did not induce adverse effects on biological functions of BMDCs.

In Vivo Imaging of BMDC Migration with the Multimodal Reporter

DCs are routinely administered via intramuscular, intravenous, and intradermal routes for DC vaccination [20], [21], [22], [23], [24]. It is important to determine whether infused DCs are successfully delivered to the target site for effective generation of antitumor immunity. Therefore, we attempted to track BMDC migration by multimodal reporter gene imaging following subcutaneous injection into the footpads of mice. Several studies have reported that small populations of DCs migrate to the draining lymph nodes, such as popliteal lymph nodes and inguinal lymph nodes, after subcutaneous injection into the footpads of mice [25], [26]. The administration of immune effectors can lead to nonspecific accumulation of radioiodine at the injection site through an immune cell–mediated inflammatory reaction. Therefore, BMDC-injected footpads were included as a control (Figure S1). It has been demonstrated that preexposure of the injection site to proinflammatory cytokines such as TNF-α results in enhanced DC migration to draining lymph nodes [27]. Therefore, in this study, mice received 30 ng TNF-α via subcutaneous injection into the hind footpads.

After injection of BMDC/NF cells, we monitored their migration with combined I-124 PET/CT and BLI on days 0 and 7 (Figure S1). On day 0 postinjection, intense BLI signals were detected in BMDC/NF-injected but not BMDC-injected footpads (Figure 6A, upper panel, red arrow). Consistent herewith, intense radioactive signals were observed only in BMDC/NF-injected footpads (Figure 6A, bottom panel, red arrow). We did not observe migrated BMDCs in DPLNs on day 0. However, on day 7 postinjection, in vivo BLI clearly showed migrated BMDCs in DPLNs (Figure 6A, upper panel, yellow arrow). Consistent herewith, in vivo PET/CT with I-124 revealed intense radioiodine signal in DPLNs (Figure 6A, bottom panel, yellow arrow). However, no BLI and radioactive signals were detected in BMDC-injected footpads. The BLI signal in DPLNs ranged from 1.04 × 105 P/cm2/s/sr to 5.62 × 105 P/cm2/s/sr (Figure 6B). There was a relative increase in radioiodine uptake by DPLNs from 17.4% to 38.2% (Figure 6C).

Figure 6

In vivo imaging of BMDC migration with the multimodal reporter construct in living mice. (A) Combined BLI (upper) and PET/CT (bottom) of the migration of BMDC/NF cells to DPLNs at days 0 and 7 postinjection. Red and yellow arrows indicate the injected site and draining popliteal lymph nodes, respectively. (B, C) Quantification of BLI and radioactive signals in footpads and DPLNs. (D) Ex vivo BLI and PET/CT of excised control popliteal lymph nodes (PLNs, left) and DPLNs (right). (E) Immunoblot analysis of excised DPLNs and control PLNs with anti-NIS and anti-luciferase antibodies to evaluate the presence of migrated BMDC/NF cells in DPLNs. Each group consisted of five mice, and experiments were performed at least in triplicate.

To identify the BMDCs that had migrated to DPLNs, the mice were sacrificed, and DPLNs were excised for ex vivo imaging and immunoblotting with anti-hNIS and anti-luc2 antibodies. Ex vivo imaging with combined BLI and PET/CT demonstrated strong signals in DPLNs but not control popliteal lymph nodes (Figure 6D), and protein expression of both hNIS and luc2 was detected in DPLNs of BMDC/NF- but not BMDC-injected footpads (Figure 6E).

These results suggested that multimodal reporter imaging with hNIS and luc2 genes is feasible for monitoring the migration of primary BMDCs to draining lymph nodes in living mice.

Application of Multimodal Reporter Gene Imaging in the Experimental Setting of DC-Based Immunotherapy

We investigated whether injection with BMDC/NF cells could successfully generate antitumor immunity in immunocompetent mice. For this experiment, we selected the mouse cervical cancer cell line TC-1 expressing E7 antigen. This model has been used to evaluate the antitumor immune response generated by DNA or DC vaccines expressing the E7 antigen [20], [28], [29]. To visualize the antitumor effects in vivo, we further engineered the TC-1 cells to express Rluc. Because Rluc uses the substrate coelenterazine, which is different from the firefly luciferase luc2 substrate, reporter BMDCs and tumor cell proliferation could be monitored simultaneously in vivo.

BMDC/NF cells were pulsed with TC-1 lysates, and mice were immunized with pulsed BMDC/NF cells once a week for 2 weeks (Figure S2). After the first intramuscular injection of pulsed BMDC/NF cells, we examined whether the cells were successfully transferred to the vaccination site using PET/CT and BLI. As shown in Figure 7A, distinct radioactive and BLI signals were detected at the injection site for both pulsed and unpulsed BMDC/NF cells, indicating successful delivery of BMDC/NF cells to the vaccination site, which is critical for the effective generation of an antitumor immune response. Similarly, we observed radioactive and BLI signals at the injection site following the second vaccination (data not shown).

Figure 7

Evaluation of antitumor immunity induction by BMDC/NF cells in a cervical cancer model by multimodal reporter gene imaging. (A) In vivo BLI and axial PET/CT imaging of unpulsed BMDC/NF cells (left) and pulsed BMDC/NF cells (right) following intramuscular injection. (B) In vivo visualization of the antitumor effects of pulsed BMDC/NF cells with Rluc reporter gene imaging. Significant tumor protection was observed in mice immunized with pulsed BMDCs. (C) Quantification of the BLI signal in tumors. (D) Measurement of tumor weight; inset: photograph of excised tumor from each group. (E) In vitro BLI imaging of the killing activity of CTLs against TC-1/Rluc ([a] splenocytes from mice immunized with unpulsed BMDCs, [b] splenocytes from mice immunized with pulsed BMDCs). Each group consisted of seven mice, and experiments were performed at least in triplicate. Values indicate means ± SDs.

After the final vaccination, the mice were challenged with TC-1/Rluc, and tumor progression was monitored with BLI (Figure S2). Until day 7 postchallenge, there were no significant differences in tumor-derived BLI signals between pulsed and unpulsed BMDC/NF groups (Figure 7C). From day 11, we observed marked tumor protection in mice immunized with pulsed BMDC/NF cells and a higher BLI signal in mice immunized with unpulsed BMDC/NF cells (P < .05). The BLI activity values were 2.3 × 1010 ± 9.8 × 109 P/cm2/s/sr and 1.2 × 1010 ± 7.1 × 109 P/cm2/s/sr in mice receiving unpulsed and pulsed BMDC/NF cells, respectively (Figure 7C). Significant tumor protection was observed in the group immunized with pulsed BMDC/NF cells until 14 days (Figure 7B), and excised tumors were smaller and weighed less in the pulsed BMDC/NF–receiving group than in the unpulsed BMDC/NF–receiving group (Figure 7D). During the experiment, we did not see any adverse effects of the therapy on body weight or fur appearance in both groups (data not shown).

To explore the immune responses in tumor-bearing mice further, we prepared splenocytes from each group to measure the killing activity of CTLs [28]. We monitored the BLI activity of TC-1/Rluc cells at an effector to target ratio of 5:1 to examine the killing activity of CTLs against TC-1 cells. As shown in Figure 7E, the BLI activity in splenocytes from mice immunized with pulsed BMDC/NF cells was lower than that in splenocytes from mice immunized with unpulsed BMDC/NF cells (P < .05).

Taken together, these results suggested that the multimodal reporter construct did not adversely affect the generation of antitumor immunity by BMDCs, while facilitating the monitoring of successful delivery following DC vaccination.

Conclusion

In summary, we developed a new multimodal reporter system, Retro-NFT, suitable for in vivo DC tracking, which allowed successful introduction of multiple reporter genes into primary BMDCs via retrovirus transduction. Retro-NFT introduction did not have adverse effects on the biological functions of these cells. Furthermore, we could monitor the migration of BMDCs in living mice by employing a combination of BLI and I-124 PET/CT. Importantly, BMDCs expressing the reporter genes could generate antitumor immunity against cervical cancer. Taken together, these results highlight the potential of our multimodal reporter gene system for application as a new imaging platform to evaluate the biological behaviors of BMDCs and to optimize DC-based immunotherapies. However, because the efficiency of viral infection of BMDCs was rather low, it was difficult to detect the early migration of BMDCs to the lymphoid organs. Therefore, it will be necessary to enhance the transfection efficiency of multimodal reporter genes in BMDCs.

Notes

The authors declare no competing financial interest.