Abstract
Background/Aim: Using the tyrosine hydroxylase (TH)-MYCN mouse neuroblastoma (NB) model, we have previously reported the accumulation of mouse mesenchymal stem cells (mMSCs) on tumors in vivo and the antitumor effect of mMSCs transfected with a small molecule (IFN-β) expression gene. In this study, we have developed novel MSCs secreting anti-disialoganglioside GD2 antibody (anti-GD2-MSCs) and evaluated their antitumor effects in vitro. Materials and Methods: We generated an anti-GD2 antibody construct (14.G2a-Fcx2-GFP) incorporating FLAG-tagged single-chain fragment variable against GD2 fused to a linker sequence, a fragment of the constant portion of human IgG1, and GFP protein. The construct was lentivirally transduced into mMSCs and the transduction efficiency was assessed by GFP expression. The secretion of FLAG-tagged anti-GD2 antibody was detected by Western blotting using anti-FLAG antibody. Antibody binding capacity was confirmed by flow cytometry. Antibody-dependent cellular cytotoxicity (ADCC) was evaluated using human NB cells and human natural killer (NK) cells to assess whether the antitumor activity was enhanced in the presence of the produced antibodies. Results: The transduction efficiency of anti-GD2-MSCs was more than 90%. anti-GD2-MSCs secreted antibodies extracellularly and these antibodies had high affinity to GD2-expressing human NB cells. ADCC assays showed that the addition of antibodies secreted from anti-GD2-MSCs significantly increased the cytotoxic activity of NK cells against NB cells. Conclusion: Newly developed anti-GD2-MSCs produced functional antibodies that have affinity to the GD2 antigen on NB cells and can induce ADCC-mediated cytotoxicity. Anti-GD2-MSCs based cellular immunotherapy has the potential to be a novel therapeutic option for intractable NB.
Cancer immunotherapy using specific antibodies targeting tumor-associated antigens has attracted substantial attention in recent years because of its ability to selectively kill tumors, minimize the non-specific impact on patients by avoiding the immune recognition of normal tissues, and achieve complete and durable remission by using the immune system to eradicate dormant cancer cells or micrometastatic sites of cancer to prevent recurrence (1). In addition, a variety of strategies to boost the immune system have been developed in recent cancer immunotherapies, including immuno-nanotechnology, such as the specific delivery of multiple immunomodulators, enhanced uptake of immunotherapy by antigen-presenting cells (APCs) and new approaches to engineering T cells. These strategies are leading to new therapies that can ameliorate the current clinical problems (2-4).
GD2 is a disialoganglioside expressed on tumors of neuroectodermal origin including human neuroblastoma (NB) and other malignancies regardless of its malignant phenotype (5, 6). Therefore, GD2 is a suitable target for immunotherapy with monoclonal antibodies. Immunotherapy with an anti-GD2 monoclonal antibody, granulocyte-macrophage colony-stimulating factor, interleukin-2 (IL-2), and isotretinoin has significantly improved survival in patients with high-risk NB (7). Although cancer immunotherapy possesses potent therapeutic efficacy, there are several problems in its clinical application due to the complex tumor microenvironment, patient heterogeneity, and systemic immunotoxicity (8, 9). Generally, cancer immunotherapy can enhance antitumor immunity against antigen-presenting tumor cells without serious side effects, but because GD2 is also expressed in normal tissues and cells including melanocytes and nerve fibers, anti-GD2 antibodies cause toxicities such as neuropathic pain, fever, and allergic reactions (1, 10). Consequently, the development of carrier-based drug delivery systems that specifically deliver antibodies to tumors is desirable in terms of reducing side effects and enhancing efficacy.
Mesenchymal stem cells (MSCs) have been identified as potential tumor-specific therapeutic delivery vehicles in NB due to their ability to homing to tumors (11, 12). MSCs are immunologically inert due to their low expression of constitutive major histocompatibility complex 1 (MHC1) and lack of MHC2 and costimulatory molecules CD80, CD86, and CD40 (13). Recent research suggests that therapeutic genes can be introduced into MSCs, which can migrate to micrometastatic sites of invasive tumors, allowing for more targeted treatment (14-17). Recombinant MSCs have also been shown to have antitumor effects on NBs (18, 19).
We previously reported that MSCs have a homing effect on tumors in the TH-MYCN mouse NB model (20, 21) and also found that mouse MSCs (mMSCs) transfected with the IFN-β gene inhibited tumor growth in the TH-MYCN mouse NB model (21). In this study, we manufactured MSC-based cellular vehicles armed to secrete anti-GD2 antibody, which were theoretically expected to facilitate specific delivery of the antibody to the tumor site and enhance anti-tumor activity with reducing systemic toxicity as a potential novel cellular immunotherapy for pediatric cancer.
Materials and Methods
Cells. Mouse MSCs (mMSCs) derived from the bone marrow of a C57BL/6 background mouse (Cyagen, Santa Clara, CA, USA) were maintained as described previously (20). The human NB cell lines IMR5 (high GD2 expression), SH-SY5Y (moderate GD2 expression), and SK-N-FI (low GD2 expression) were maintained in RPMI 1640 (Nacalai Tesque, Kyoto, Japan) supplemented with 10% fetal bovine serum with 1% penicillin/streptomycin (Nacalai Tesque). The cells were incubated at 37°C in 5% CO2 and 95% humidity. All NB cell lines were kindly provided by the laboratory at the Children’s Hospital of Philadelphia (PA, USA) and passaged in our department.
GD2 antibody construct design and lentiviral transduction. GD2-specific single-chain fragment variable (scFv14.G2a), which was previously identified and applied in clinical trials (22-24), was fused with human IgG1 hinge and two Fc regions (one of which was FLAG-tagged) by a Gly–Gly–Gly–Ser linker connection. The created construct of the cDNA of FLAG-tagged single-chain anti-GD2 antibody was named “14.G2a-Fcx2-FLAG”. 14.G2a-Fcx2-FLAG was cloned into the EcoRI/NheI-digested pCDH-EF1-MCS-T2A-copGFP, an HIV-based self-inactivating lentiviral vector plasmid (System Biosciences, Palo Alto, CA, USA). The lentiviral plasmid 14.G2a-Fcx2-GFP (Figure 1A) and the packaging plasmid were co-transfected into 293T cells and cultured. Viral supernatants were concentrated and transduced into mMSCs. Production and enrichment of viral supernatant and transduction into mMSCs were performed using the same procedures as previously reported (25). Anti-GD2-MSCs were photographed after transfection using a BZ-X710 microscope (Keyence, Osaka, Japan). The fluorescence intensity of transfected cells was observed using a FACSCalibur (Becton, Dickinson and Company, Franklin Lakes, NJ, USA).
Composition of the introduced anti-GD2 antibody constructs and evaluation of the reporter GFP expression of anti-GD2 mesenchymal stem cells. (A) Construction of the anti-GD2 antibody (14.G2a-Fcx2-GFP). (B) GFP expression in MSCs transduced with lentiviral constructs carrying anti-GD2 antibody and GFP was assessed by fluorescence microscopy (scale bar: 100μm). (C) GFP expression in anti-GD2-MSCs 7 days after transduction was analyzed by flow cytometry. (D) Anti-GD2-MSCs were cultured and analyzed by flow cytometry at different intervals after transduction to determine the percentage of GFP-positive cells. Data are presented as the mean±standard deviation of triplicate samples. Anti-GD2-MSCs, Anti-disialoganglioside GD2 antibody-secreting MSCs; LTR: long terminal repeat; EF1: elongation factor1; scFv, single-chain variable fragment.
Western blotting. MSCs and anti-GD2-MSCs were cultured at a density of 1×106 cells per 100-mm dish plate, and their supernatants were collected after 72 h. Each sample was treated with a Spin column-based Antibody Purification Kit (Protein G, Cosmo Bio, Tokyo, Japan). Proteins were detected from prepared samples (anti-GD2-MSCs-sup, MSCs-sup) and FLAG peptides (control) as previously reported using the following (25): polyvinylidene chloride membrane (Bio-Rad, Hercules, CA, USA), Blocking One (Nacalai Tesque), monoclonal ANTI-FLAG® M2 antibody (1:1000; MilliporeSigma, Merck KGaA, Darmstadt, Germany) and mouse IgG HRP-conjugated antibody (1:10,000; R&D Systems, Minneapolis, MN, USA). Immunoreactive bands were visualized using ImageQuant LAS500 (Cytiva, Tokyo, Japan).
Detection of GD2 expression in NB cell lines and the affinity assay for antibodies secreted by anti-GD2-MSCs. GD2 expression in human NB cell lines was detected using anti-GD2-PE, clone 14.G2a (cat no. 357304, BioLegend, San Diego, CA, USA). PE mouse IgG2a, 
isotype control (FC) antibody (cat no. 400213, BioLegend) was used as the isotype control. IMR-5, SH-SY5Y and SK-N-FI cells were incubated with MSCs-sup or anti-GD2-MSCs-sup for 30 min and then used for the affinity assay. The binding of FLAG-tagged human IgG1 derived anti-GD2 antibody onto the surface of NB cells treated with each supernatant was detected using APC anti-DYKDDDDK Tag antibody (cat no. 637307, BioLegend) and FITC anti-human IgG Fc antibody (cat no. 410719, BioLegend). Data were analyzed using the Cell Quest software (BD Bioscience).
Extraction of peripheral blood mononuclear cells and NK cells. Peripheral blood mononuclear cells (PBMCs) were isolated from the heparinized venous blood of healthy volunteers by density gradient centrifugation using Ficoll-Paque PLUS (Cytiva). Human NK cells were isolated from PBMCs using an NK isolation kit (Miltenyi Biotec, GmbH, Bergisch Gladbach, Germany). The procedure for the use of NK cells in this study was approved by the Ethical Committee of Kyoto Prefectural University of Medicine (Permission number: ERB-C-2012).
Antibody-dependent cellular cytotoxicity (ADCC) assay. NK cell-mediated cytotoxicity was analyzed by flow cytometry using a PKH26 red fluorescent cell linker mini kit for general cell membrane labeling (Millipore Sigma, Merck KGaA) and a 7-aminoactinomycin D (7-AAD) cell viability staining (BD Biosciences, Becton, Dickinson, and Company). IMR-5, SH-SY5Y and SK-N-FI cells stained by PKH26 were plated at a density of 50,000 cells per well with NK cells in a 24-well plate and incubated with anti-GD2-MSCs-sup or MSCs-sup containing recombinant human IL-2 (R&D Systems) at a concentration of 100 IU/ml in each well. Assays were performed at NK cell:NB cell [effector:target (E:T)] ratios of 0:1 (untreated) to 10:1. After co-culture for 4 h, cells were harvested, 7-AAD was then added (0.25 μg/ml final concentration), and cells were analyzed by flow cytometry. The percentage of target cell death was corrected for spontaneous background death by subtracting the percentage of dead cells in control samples (PKH-26–labeled target alone) from the percentage of dead cells within test samples.
Statistical analyses. The experiments were performed in triplicate. Statistical significance was determined using Tukey’s multiple comparison test. Data were considered statistically significant at p<0.05. Data are expressed as the mean±standard deviation. All statistical analyses were performed using EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan).
Results
Evaluation of transfection efficiency by detecting GFP expression using fluorescence microscopy and flow cytometry. We initially introduced the transgene into MSCs using the CMV promoter, but transgene expression was minimal, probably due to transcriptional repression by gene silencing. Therefore, we chose the EF1a promoter (26). Almost all of the generated anti-GD2-MSCs expressed GFP, and the transfection efficiency exceeded 90% (Figure 1B-C). Expansion and passaging did not change the percentage of GFP-expressing cells (Figure 1D).
Anti-GD2 antibodies were secreted extracellularly by anti-GD2-MSCs. Antibodies secreted by anti-GD2-MSCs carried a marker peptide (FLAG) at the C-terminus. The presence of FLAG-tagged antibodies in the anti-GD2-MSC culture supernatant was confirmed by Western blotting of the samples extracted from anti-GD2-MSC or MSC culture supernatant and purified using an IgG column (Figure 2). The molecular weights indicated by the detected bands were also as expected (approximately 90 kDa). The MSC supernatant did not contain FLAG-tagged IgG.
Western analysis of FLAG protein (positive control), MSCs-supernatant (negative control), or Anti-GD2-MSCs-supernatant. The molecular weights of the FLAG-tagged anti-GD2 antibody and FLAG protein are approximately 90 and 64 kDa, respectively. Anti-GD2-MSCs, Anti-disialoganglioside GD2 antibody-secreting MSCs.
Antibodies secreted by anti-GD2-MSCs exhibited affinity for GD2 antigen on the surface of neuroblastoma cells. GD2 expression in three human NB cell lines is shown in Figure 3A. The fluorescently labeled anti-GD2 monoclonal antibody used for flow cytometry in Figure 3A has an antigen-binding site derived from 14.G2a, similar to our construct. These NB cell lines were used to check whether antibodies secreted by anti-GD2-MSCs have affinity for the GD2 antigen.
GD2 antigen expression in human neuroblastoma cell lines and the binding capacity of anti-GD2 antibody secreted by anti-GD2 mesenchymal stem cells (MSCs). (A) The surface expression of GD2 on three neuroblastoma cell lines was evaluated by fluorescence-activated cell sorting. A GD2 isotype antibody was used as a negative control for detecting GD2 expression (light gray histograms). IMR-5, SH-SY5Y, and SK-N-FI and cells exhibited high, intermediate, and low GD2 expression, respectively (empty histograms). (B) Three neuroblastoma cell lines were suspended in the supernatant of anti-GD2-MSCs or MSCs and incubated at 4°C for 1 h. FLAG (left) and human IgG (right)-Fc portions of anti-GD2 antibody bound to the surface of neuroblastoma cells were detected by flow cytometry. Neuroblastoma cells incubated with non-transduced MSC supernatant are presented as light gray histograms, and neuroblastoma cells incubated with anti-GD2-MSC supernatant are presented as empty histograms. Anti-GD2-MSCs, Anti-disialoganglioside GD2 antibody-secreting MSCs.
Each NB cell line was suspended and incubated in anti-GD2-MSC or MSC supernatant, and the Fc region and FLAG tag on the surface of tumor cells were evaluated by flow cytometry. The Fc region and FLAG were identified on the surface of all three NB cell lines reacted with anti-GD2-MSC supernatant according to the intensity of GD2 expression (Figure 3). These data indicated that antibodies secreted by anti-GD2-MSCs could bind to human NB cells.
Antibodies secreted by anti-GD2-MSCs increased NK cell-induced cytotoxicity. Next, we tested whether the antibodies secreted by anti-GD2-MSCs caused ADCC, which is a major role of anti-GD2 antibody for neuroblastoma. There was no change in the relative viability of any NB cell line in the antibody-only group (anti-GD2-MSC culture supernatant without NK cells) compared to that in the control group (MSC culture supernatant without NK cells; Figure 4). Strikingly, the relative viability was significantly decreased in IMR5 (GD2-high) in the presence of antibodies (anti-GD2-MSC culture supernatant), compared to that in the absence of antibodies (MSC culture supernatant), when NK cells were present at E:T ratios of 1:1 (p=0.039) and 5:1 (p=0.005). At an E:T ratio of 10:1 (p=0.532), cytotoxicity tended to be higher in the presence of antibodies, although the difference was not significant (Figure 4A). The relative viability of SH-SY5Y (GD2-mid) cells was significantly lower in the presence of antibodies only for an E:T ratio of 5:1 (p=0.001). Cytotoxicity tended to be higher in the presence of antibodies at E:T ratios of 1:1 (p=0.608) and 10:1 (p=0.317), albeit without significance (Figure 4B). In SK-N-FI (GD2-low) cells, viability was not altered in the presence of antibodies for any E:T ratio (Figure 4C). These results indicate that anti-GD2-MSC culture supernatant containing anti-GD2 antibodies increases the cytotoxic effects of NK cells against NB cells. In addition, we conducted non-contact co-culture experiments with anti-GD2-MSCs and NB cells. After 72 hours of culturing IMR5 (GD2-high) cells in anti-GD2-MSC supernatant, MSC supernatant, and normal culture medium, the number of tumor cells cultured in anti-GD2-MSC supernatant was lower than the others. The results suggested that anti-GD2-MSCs had no effect on promoting NB cell growth in vitro (Supplementary Figure).
Antibody-dependent cellular cytotoxicity (ADCC) of anti-GD2 antibody secreted by anti-GD2 mesenchymal stem cells (MSCs). Natural killer (NK) cell-mediated ADCC induced by anti-GD2 antibody from anti-GD2-MSCs was evaluated via co-culture with three neuroblastoma cell lines at different effector:target (E:T) ratios. Relative survival rate of each cell line. NK cells were cultured for 4 h with neuroblastoma cells in MSC supernatant (white) or anti-GD2-MSC supernatant (dark gray) with the interleukin-2 concentration adjusted to 100 IU/ml. Error bars indicate the standard errors of the means of three experiments. *p<0.05, as determined by Tukey’s t-test. anti-GD2-MSCs, Anti-disialoganglioside GD2 antibody-secreting MSCs.
Discussion
In this study, we demonstrated that anti-GD2-MSCs, which were genetically modified to produce anti-GD2 antibodies, exhibited anti-tumor efficacy via ADCC mediated by NK cells. Therefore, anti-GD2-MSCs could represent a promising therapeutic product that can enhance drug delivery to tumor sites and reduce unwanted systemic off-tumor toxicity, potentially caused by on-target cross-reactivity to these antibodies with normal tissues. However, the possibility that anti-GD2 antibodies counteracted the proliferation-promoting effect of MSCs cannot be dismissed because it has been suggested that anti-GD2 antibodies directly induce cell death in GD2-positive tumor cells without activating the immune mechanism (27, 28).
We previously demonstrated that mMSCs expressing IFN-β could migrate into the intra-abdominal tumors of TH-MYCN transgenic mice and exert an anti-tumor effect (21). However, the anti-tumor efficacy was transient, and tumor growth was not controlled. Because IFN-β is not a standard treatment for NB despite its ability to induce apoptosis, we chose anti-GD2 antibodies, which have already been applied clinically and which have validated efficacy against NB, as the therapeutic agent to be equipped for MSCs.
The current issues associated with clinical anti-GD2 antibody therapy are the immunosuppressive tumor microenvironment and low tumor penetration of antibodies (29). However, the most challenging aspect of this therapy is the side effect of neuropathic pain (30), which is caused by maintaining the serum antibody levels, leading to dose limitation. In this respect, anti-GD2-MSCs are expected to reduce the side effects as well as the cost due to the ease of preparation. In this context, anti-GD2-MSCs accumulated in the tumor, as demonstrated by the in vitro assay, leading to the local secretion of the GD2 antibodies at the tumor site, which could induce NK cell-mediated ADCC in NB cells. Although we did not demonstrate the possibility that anti-GD2-MSCs invaded normal tissues including GD2-positive neurons, the homing ability of MSCs to tumor sites would circumvent off-tumor reactions in normal tissues, although this hypothesis should be definitely validated using in vivo syngeneic models. Nevertheless, this is the first study to prove that MSCs delivered to tumor sites can produce therapeutic antibodies and cause ADCC.
Importantly, anti-GD2-MSCs caused NK cell-mediated ADCC, which is the main mode of action of anti-GD2 antibodies rather than direct anti-tumor activity of the antibodies. As expected, ADCC was not induced in SK-N-FI cells, suggesting that cells with low GD2 expression including normal tissue would not be affected by anti-GD2-MSC–induced cytotoxicity.
The concept of anti-GD2-MSCs as a cellular therapy is promising, and these cells indeed exhibited tumor control. However, there remain limitations in applying this approach in the clinic. Although the homing of MSCs to tumors has been extensively investigated, the anti-tumor ability of anti-GD2-MSCs to avoid indiscriminate interactions with normal tissues should be validated using in vivo models. In addition, this experiment was validated in the presence of sufficient amounts of anti-GD2-MSCs to maintain antibody concentrations sufficient to cause ADCC. Further experiments are needed to determine whether sufficient local antibody concentrations can be maintained in vivo. Furthermore, it is possible that untransfected, non-armed MSCs would enhance tumor growth, as reported in several studies. Recent studies have shown that MSCs interact with and influence tumor cells at various stages of disease progression. MSCs increase the metastatic potential of tumor cells by promoting their motility and invasiveness as well as play a role in the formation of a metastatic niche at the secondary site (31-33); however, anti-tumor effects were observed in other studies, demonstrating their ability to inhibit the growth and metastasis of various malignancies (34-40). Because the tumor-promoting properties of MSCs may be partly related to their immunomodulatory effects (41), MSCs can impair the proliferation of T cells at the tumor site, the release of pro-inflammatory cytokines by dendritic cells and macrophages, B cell proliferation and immunoglobulin production, and the cytotoxic activity of NK cells (42). Nevertheless, we demonstrated the ability of anti-GD2-MSCs to induce NK cell-mediated ADCC without impairing NK cell function, which should be validated further using in vivo syngeneic models. As a next step of the present study, we are currently validating the efficacy and safety profiles of this treatment for in vivo mouse models using a modified anti-GD2 antibody with a fragment of a stationary portion changed from human IgG to mouse IgG and GFP. The preliminary data for syngeneic localized NB mouse model has just been published (25), and further experiments including an osteosarcoma model are now under planning.
Conclusion
In conclusion, we developed anti-GD2-MSCs as a potential novel cellular immunotherapy for NB. In this study, we generated a construct that is easy to evaluate for transduction efficiency and antibody secretion after gene transfer and introduced it into MSCs. The efficiency of the transfection was more than 90%, and anti-GD2-MSCs secreted FLAG-tagged anti-GD2 antibodies extracellularly as designed. Similar to general anti-GD2 monoclonal antibodies, antibodies secreted from anti-GD2-MSCs had affinity for GD2 antigen and ADCC activity via NK cells. Because anti-GD2-MSCs have the potential to specifically kill tumor cells without affecting normal cells with weak GD2 expression, anti-GD2-MSC–based cellular immunotherapy represents a novel therapeutic option for intractable NB.
Acknowledgements
We are grateful to the members of the Department of Immunology, Kyoto Prefectural University of Medicine for useful comments. The procedure for the use of NK cells in this study was approved by the Ethical Committee of Kyoto Prefectural University of Medicine (Permission number: ERB-C-2012). This work was supported by a Grant-in-Aid for Exploratory Research from the Ministry of Education, Culture, Sports, Science and Technology of Japan (MEXT KAKENHI grant number 19H03719) and by the Practical Research for Innovative Cancer Control from the Japan Agency for Medical Research and Development, AMED (grant number 20ck0106609h0001). The authors would like to thank Enago (www.enago.jp) for the English language review.
Footnotes
Authors’ Contributions
Conceptualization: MI, SY, TT. Investigation: MI, KK, MH, SF. Validation: MI, SY, MH, TK. Supervision: SY, SF, TI, OM, TT. Writing – original draft: MI. Writing – review and editing: MI, SY, KK, MH, SF, TK, TI, OM, TT. Resources: SY, SF, TI, TT. Funding acquisition: SF, TT. All Authors contributed to manuscript revision, read, and approved the submitted version.
Supplementary Material
Supplementary data in this paper are available as open data through the online data repository of the Department of Pediatric Surgery, Kyoto Prefectural University of Medicine at:<http://pedsurg.kpu-m.ac.jp/system/upload/pdf/Supplemental%20Figure.pdf>
Conflicts of Interest
The Authors have no conflicts of interest directly relevant to the content of this article.
- Received March 23, 2023.
- Revision received April 15, 2023.
- Accepted April 19, 2023.
- Copyright © 2023 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.