Immunotherapy is an effective and extremely potent modality for treating various cancers due to its long-lasting effects. Use of blocking antibodies to target programmed death 1 (PD-1) and its ligand (PD-L1) has dramatically changed the landscape of therapies for multiple cancers.1,2 Although durable antitumor responses have been observed in some patients receiving cancer immunotherapy, the benefit of immune checkpoint inhibitors still exerts a limited effect on survival in patients with certain types of cancer.3 Therefore, therapeutic strategies based on overcoming immune resistance are urgently needed to enhance anticancer immunity.4–6
Siglecs are a family of sialic acid immunoglobulin receptors that play important roles in tumor immunosurveillance making them attractive targets for cancer treatment.7 Siglec-15 was identified as a member of the Siglecs family. Its extracellular domain consists of only 1 IgV and 1 IgC2 domain, and it exhibits 20%–30% identity with the B7 family.8,9 Previous studies on Siglec-15 have focused on its function in osteoclast differentiation and bone remodeling, and Siglec-15 is considered to be a potential therapeutic target for postmenopausal osteoporosis.10–19 Recently, Siglec-15 was identified as a novel immunosuppressive molecule that plays an important role in tumor immunity.
Siglec-15 predominantly binds to the tumoral sialyl-Tn structure, and is broadly upregulated in cancer cells and tumor-infiltrating macrophages/myeloid cells; moreover, it has been significantly associated with poor outcomes for patients with certain types of cancer.12,20–28 Targeting Siglec-15 is viewed as a promising therapeutic approach for the development of cancer immunotherapies, which may be used to complement first-generation immunotherapies against cancer. Recently, targeting Siglec-15 with monoclonal antibodies (mAbs) has shown therapeutic effects in several murine cancer models.20,21,26,29 Notably, the phase I clinical trial results of a humanized mAb (NC318) used against advanced non–small cell lung cancer (NSCLC) led to a promising clinical response. A phase II clinical trial for evaluating NC318 alone and in combination with pembrolizumab in advanced NSCLC patients is ongoing (phase II study: NCT04699123).
An anti-Siglec-15 mAb plays a significant role in tumor killing via various mechanisms; for example, the mAb reversed Siglec-15-mediated suppression of T cells and disrupted the interaction between Siglec-15 and its ligand; however, the mechanism underlying anti-Siglec-15 mAb effects on cancer remains unclear. In this study, we initially identified a novel anti-Siglec-15 mAb using hybridoma technology. In particular, this mAb showed many excellent properties, such as high affinity, species cross-reactivity, and greater reversal of Siglec-15-mediated inhibition of T-cell activity. The essential role of Fc-mediated effector functions of several other immune modulator mAbs has been reported in the clinic.30–32 However, the Fc effector function of anti-Siglec-15 mAb has not been explored. Hence, we developed a novel mAb and explored the contribution of Fc-mediated effector functions to the activity of the anti-Siglec-15 mAb in vivo.
MATERIALS AND METHODS
Animals
All animals have been maintained in the local specific pathogen free animal facility with a 12/12 hours light/dark cycle and mouse chow and water ad libitum. This study was performed in strict accordance with institutional guidelines and approved by the Institutional Animal Care and Use Committee of the Shanghai Model Organisms Center Inc. and the IACUC permit number was 2019-0011. Balb/c mice were purchased from Charles River, and Siglec-15 humanized mice were provided by Shanghai Model Organisms. Regular mice were used for experiments at the age of 5–8 weeks. A maximum tumor size exceeding 3000 mm3 was defined as a humane endpoint. This study was performed in strict accordance with institutional guidelines and approved by the Institutional Animal Care and Use Committee.
Cell Culture
SP2/0 cells were grown in RPMI 1640 supplemented with 10% (v/v) fetal bovine serum (FBS). The Hek293 cells were grown in DMEM supplemented with 10% (v/v) FBS. U87-MG cells were cultured in MEM supplemented with 1% NEAA and 10% FBS. All these cells were obtained from ATCC and cultured at 37°C in the presence of 5% CO2 in a humidified incubator.
Generation of Stable Cell Lines
Stable cell lines overexpressing human Siglec-15 (uniprot: Q6ZMC9) and Leucine-rich repeat-containing protein 4C (LRRC4C) (uniprot: Q9HCJ2) were generated by transfecting Hek293 cells with the respective gene cloned into the pcDNA3.1(+)vector（GenScript). The constructed plasmids were transfected into Hek293 cells using Lipofectamine 3000, respectively. Stable cell lines expressing human Siglec-15 (Hek293-hsiglec-15) or LRRC4C (Hek293-LRRC4C) were selected in the presence of G418 (700 µg/mL, Sigma-Aldrich). MC38 cells stably expressing human Siglec-15 were generated using a lentiviral system for transduction (the pSLenti-EF1-Luc2-F2A-puro-CMV-SIGLEC-15-HA-WPRE lentiviral system was purchased from OBiO). Briefly, MC38 cells were seeded in a 24-well plate at a density of 5Ă—104 cells/well, and transduced with 20 MOI lentiviral particles with 5 μg/mL Polybrene for 17 hours. Forty-eight hours later, the cells were selected with 2 µg/mL puromycin for 3 days. Single cells were then plated in individual wells of a 96-well plate for 10 days. Siglec-15 expression was checked by FACS using anti-Siglec-15 mAb 1-10E5 (this work). LRRC4C expression was checked by FACS using Siglec-15-Fc protein.
Animal Immunization, Hybridoma Fusion, and Screening
A total of 6 Balb/c mice were immunized with recombinant human extracellular Siglec-15 (Fc tag) fragments at days 1, 7, and 14. The splenic cells were fused with the SP2/0 cell line at a ratio of 5:1–3:1 by electrofusion. Hybridomas were maintained in RPMI 1640 supplemented with 10% FBS and hypoxanthine-aminopterin-thymidine (HAT, Sigma-Aldrich) medium in 96-well microtiter plates for 8–10 days. Then, we changed the medium to hypoxanthine-thymidine (HT, Sigma-Aldrich) medium for an additional 3–5 days. The supernatant from each well was subjected to an indirect enzyme-linked immunosorbent assay (ELISA) using a recombinant human extracellular Siglec-15 (His tag) fragment. Positive hybridoma clones were then selected and further tested for their activities.
Indirect ELISA
To detect the antigen-specificity and cross-reactivity of secreted antibodies, an indirect ELISA was performed. Briefly, soluble Siglec-15 was coated on an ELISA plate and incubated overnight at 4°C. After 2 hours of blocking, the supernatant from each well or different concentrations of antibodies (from 0 to 5000 ng/mL) were added to the plate and incubated for 2 hours at 37°C. Then, horseradish peroxidase-conjugated goat anti-mouse IgG (Jackson ImmunoResearch #115-035-071) was added and incubated for 1 hour at 37°C. The unbound protein or antibody was washed off at every coating step with phosphate buffered saline with tween-20. 3,3′, 5,5;-tetramethylbenzidine peroxidase was used as the horseradish peroxidase substrate and then terminated with 2 M H2SO4. Finally, the absorbance at 450 nm was determined.
Flow Cytometric Assay
Hek293-hsiglec-15 or U87 cells were harvested and incubated with anti-Siglec-15 mAbs (from 0 to 10,000 ng/mL or 0 to 45,000 ng/mL) for 1 hour at 4°C. After incubation, the samples were washed with PBS 3 times and stained with Alexa488-conjugated anti-human (Invitrogen #A-11013) or mouse (Invitrogen #A32723) secondary antibodies in the dark for 50 minutes at 4°C. The mean fluorescence intensity of cells was measured by FACScan (Thermo). The data were analyzed using FlowJo software.
Cell-based Blocking Assays
Briefly, the hsiglec-15-Fc protein was incubated with anti-Siglec-15 mAbs or isotype mIgG (Biointron #B115101) on ice for 30 minutes, respectively, followed by incubation with Hek293-LRRC4C cells (2×105 cells/sample) on ice for 50 minutes. Cells were then washed with PBS to remove the unbound hsiglec-15-Fc. After washing, samples were stained with Alexa488-conjugated anti-human secondary antibodies (Invitrogen #A-11013) in the dark for 40 minutes at 4°C. The fluorescently labeled cells were measured by FACScan (Thermo). The data were analyzed using FlowJo software.
BLI Assays
Binding affinity measurements were performed using the ForteBio Octet RED96 System. Anti-Penta-HIS or Protein A capture biosensors (ForteBio) were used to immobilize protein or mAb. Then, the sensors were washed with kinetic buffer for 120 seconds and reacted with various concentrations (from 1.56 to 50 or 100t nM) of mAbs for the 120 of mAbs for or protein A capture biosensors (ForteBio) were used to immobilize protein or mAb. Then, the sensors were washed with kinetic buffer for 120 seconds and reacted with various concentrations (from 1.56 t, and the equilibrium dissociation constant (KD) was calculated using Octet Data Analysis software (ForteBio).
IFN-γ Release and Cell Growth Assays
Total PBMCs (AllCells) from a healthy human donor were resuspended in cell culture media (RPMI1640 + 10% FBS) at a density of 150,000 cells per well in triplicate. Meanwhile, the anti-CD3 mAb (OKT3, Invitrogen #16-0037-81) was added to the PBMCs at an optimized concentration (50 ng/mL). Siglec-15-Fc protein (20 μg/mL) plus the mAbs or the isotype control hIgG (Biointron #B117901) were premixed before being added to the stimulated PBMCs. After the mixture was added to the PBMCs, the plate was incubated at 37°C and 5% CO2 for 24 hours, and the level of IFN-γ in the cell culture supernatant was quantified using a Cisbio HTRF kit. For the cell growth assay, the Siglec-15 protein and mAb mixture were incubated at 37°C and 5% CO2 for 72 hours. Cell viability was measured using the CellTiter Glo kit, according to the manufacturer’s protocol.
Receptor Internalization in Hek293-hsiglec-15 Cells
Receptor internalization was quantitatively assessed using FACS. Hek293-hsiglec-15 cells were collected and seeded at 3–6Ă—105 cells per well in a 6 well plate. Antibodies (10 μg/mL) were incubated with cells for 1 hour on ice to enable specific binding to the cell surface targets. After incubation, the cells were washed 3 times with cold PBS to remove unbound antibodies. The cells were then incubated at 37°C for 20 hours with or without antibodies to allow endocytosis to occur. Alexa488-conjugated secondary antibody was used to detect the binding of antibodies to the cells. The mean fluorescence intensity of cells was measured by FACScan (Thermo). FlowJo software was used to analyze the flow cytometry data.
pHAb-labeled Receptor Internalization
Anti-Siglec-15 mAbs were conjugated with pHAb amine-reactive dye (Promega #G9841) following the manufacturer’s instructions. Cells were trypsinized and plated in a 6-well plate at a density of 3–6Ă—105 cells per well for 20–24 hours before treatment with pHAb-labeled antibodies. Then, pHAb-conjugated antibodies were added to the cells (10 μg/mL) for 1 hour at 4°C. After incubation, the cells were washed 3 times with PBS to remove unbound antibodies. The cells were then incubated at 37°C for 20 hours to allow internalization. Hek293 cells were used as control cells. Cell images were acquired with fluorescence microscopy.
Generation and Identification of Transgenic Mice
Humanized B-Siglec-15 transgenic mice were generated using CRISPR-Cas9 technology. The Cas9 mRNA, sgRNA (gRNA1:5′-GAAAACTAGAAGAGACGCTTCGG-3′; gRNA2:5′-GAGCGCTAGGGTCGAGGTTCCGG-3′) and donor vector were microinjected into the fertilized eggs of C57BL/6J mice to generate F0 generation mice. The positive F0 generation mice identified by PCR amplification and sequencing were mated with C57BL/6J mice. For PCR analysis, genomic DNA was extracted from tails, and amplified using primers (5′ arm: Forward 5′-CAAGCCCTCTCCCCCTTTGG-3′, Reverse 5′-GGACTATTCCCCTCTTCGTTTCCTAA-3′; 3′ arm: Forward 5′-ACCCAAGTGCCCATTCCTC-3′, Reverse 5′-TCCCAGACATCCCTACCTCCAA-3′) that produced two 3.7 kb fragments from mice carrying the transgene and confirmed by DNA sequencing.
Tumor Model
Siglec-15 humanized C57BL/6 mice were subcutaneously (s.c.) injected with 3Ă—106 MC38-hS15 cells. Six days later when the tumors reached ~80–100 mm3, mice were staged and randomized into the indicated groups and dosed intraperitoneally twice a week with mAbs or normal saline. Tumor growth was monitored by calipers and calculated using the equation (length Ă— width2)/2. Antitumor efficacy was indicated by the endpoint tumor growth inhibition (TGI) rate. The TGI rate was calculated using the formula TGI (%) = (1−dT/dC) Ă— 100, where dT is the final tumor volume minus the beginning tumor volume of the treatment group and dC is the final tumor volume minus the beginning tumor volume in the control group.
Epitope Binning Assay
For the epitope binning experiments, hsiglec-15-his protein (5 μg/mL) was immobilized onto the anti-Penta-HIS sensors for 200 seconds. Immediately, the sensors were exposed to the first antibody for 180 seconds to saturate all epitopes of the Siglec-15 protein. Finally, the biosensors were exposed to secondary antibodies under the same conditions. When the pair of antibodies was classified in the same epitope bin (competitor), no additional binding was observed with the second antibody binding signal of the first antibody. In contrast, if additional binding by the second antibody was observed over the binding signal of the first antibody, the pair of antibodies was classified in unoccupied epitope bins (noncompetitor). Raw data were processed using ForteBio’s Data Analysis Software 12.1.
Statistical Analysis
Statistical analysis was performed using Prism 6.0 (GraphPad Software). Statistics are presented as the mean ± SEM (shown in error bars) of at least 3 independent experiments. Statistical significance was defined as P values <0.05 which were calculated using the Student t test in Microsoft Excel software (*P < 0.05, **P < 0.01, ***P < 0.001).
RESULTS
Production and Characterization of Anti-Siglec-15 mAbs
We generated a single substitution mutation, R143A, in Siglec-15 to facilitate mAb screening (Fig. 1A), since the IgV domain and residue R143 of Siglec-15 are sufficient to mediate sialylated glycan binding.8,33 To generate anti-Siglec-15 mAbs, mice immunized with high-affinity antibodies were selected for the cell fusion experiment. After 2 rounds of selection by ELISA, the best candidate was identified based on its high binding affinity for Siglec-15. One mAb (designated as 10E5, this work) was selected as the control for the subsequent study. Using ELISA with the Siglec-15-R143A mutant coated on plates, we found that 1-15D1 bound weakly to the R143A mutation protein, whereas 1-10E5 bound to the protein at concentrations ranging from 30 to 240Â ng/mL (Figs. 1B, C). To measure the titer of 1-15D1, ELISA were performed with recombinant soluble extracellular human Siglec-15 coated on plates. The EC50 value of 1-15D1 was 4.98Â ng/mL (Fig. 1D).
;)
FIGURE 1: Characterization and cross reactivity of anti-Siglec-15 monoclonal antibodies. A, Diagram of human Siglec-15. B and C, Solid-phase ELISA analysis of 1-15D1 (black) and 10E5 (gray) bound to coated human Siglec-15 (B) or Siglec-15-R143A mutation (C). D–F, Cross reactivity was detected by solid-phase ELISA of 1-15D1(black) and 10E5 (gray). Binding assay to human (D), cynomolgus monkey (E) and mouse (F) Siglec-15. Data are represented as the mean ± SEM (shown in error bars) for 3 independent experiments in triplicate. C indicates C-terminus; IgC2, a constant type 2 domain; IgV, immunoglobulin variable domain; N, N-terminus; R143A, Siglec-15 mutant R143A; TM, transmembrane domain.
Furthermore, the potential species cross-reactivity of 1-15D1 against mouse and cynomolgus monkey Siglec-15 was investigated using protein from each species. The results showed that 1-15D1 bound only to monkey Siglec-15 (EC50 9.08Â ng/mL) (Figs. 1E, F). In summary, we initially developed a mAb that bound to the extracellular domain of both human and monkey Siglec-15 with high affinity and that selectively recognized the Siglec-15 and R143A mutant proteins.
1-15D1 Recognizes Cancer Cell Membrane-anchored Siglec-15 and Blocks the Interaction Between LRRC4C and Siglec-15
To evaluate whether 1-15D1 recognizes cell surface human Siglec-15, we constructed a Hek293 cell line that stably expressed human Siglec-15. Using flow cytometry, we validated the high binding affinity of 1-15D1 to cell surface Siglec-15, which showed an EC50 of 339Â ng/mL (Fig. 2A). Next, the binding affinity of 1-15D1 for U87 glioblastoma cells was evaluated by flow cytometry.21 A binding curve was generated on the basis of 7 concentrations ranging from 0.8 to 45,000Â ng/mL and an EC50 value of 310Â ng/mL (Fig. 2B). These results indicated that 1-15D1 recognized human membrane-anchored Siglec-15.
FIGURE 2: Monoclonal antibody (mAbs) recognizes cell membrane-anchored Siglec-15 and blocks the interaction between LRRC4C and Siglec-15. A and B, Binding of 1-15D1 to human Siglec-15 overexpressing Hek293 stable cells at a concentration range from 20.6 to 10,000 ng/mL (A) and U87 cells (B) by FACS. C, Hek293 cell line that stably expressed LRRC4C was incubated with anti-Siglec-15 antibodies (20 μg/mL) and purified Siglec-15-Fc protein (2 μg/mL). Histogram (upper) and summary graph (lower) for untreated cells (deepgray), and 1-15D1-treated cells (lightgray) were shown. D, Biacore Ă—100 sensorgrams were obtained for the binding of 1-15D1 and the positive control 5G12 to immobilized Siglec-15. The mAbs were injected at 6 different concentrations ranging between 1.56 and 50 nM (shown in different colors). The negative control was hIgG. Data analysis and fitting were performed using ForteBio’s Data analysis software version 12.1.
It had been previously demonstrated that Siglec-15 interacts with several binding partners, such as myelin-associated glycoprotein, LRRC4C, and sialyl-Tn.34 The candidate antibody 1-15D1, which had been purified from the supernatant of hybridoma cells, was subsequently analyzed via a competitive FACS assay to determine its ability to competitively block soluble Siglec-15 binding to human LRRC4C that was stably expressed on Hek293 cells. The results demonstrated, 1-15D1 exhibited nearly 90% inhibition of recombinant Siglec-15 binding to cell surface LRRC4C (Fig. 2C).
Chimera generation is the preliminary step before antibody humanization and the combination of rodent variable domains with human constant domains is the most common method used to produce chimeras. Therefore, we obtained the variable domain sequence of 1-15D1 (Supplementary Digital Content Table 1, Supplemental Digital Content 1, https://links.lww.com/JIT/A714). Next, we constructed a chimeric antibody that contained a variable domain of 1-15D1 and a constant region of human IgG1 with the N297A mutation. Bio-Layer Interferometry (BLI) was performed to characterize the affinities of chimeric 1-15D1 and the positive control 5G12 (anti-Siglec-15 mAb, same sequence as reported)21 for the human Siglec-15. The results showed that 1-15D1 bound Siglec-15 with stronger affinity than 5G12, with a KD value of <10−3 nM versus that of 0.26 nM (Fig. 2D). The difference was exclusively attributed to the rate for kinetic dissociation constants (koff) of <1E7 s−1 and 1.78E-4 s−1, respectively (Supplemental Digital Content Table 2, Supplemental Digital Content 2, https://links.lww.com/JIT/A715).
1-15D1 Reverses Siglec-15–mediated Suppression of Human PBMCs and Promotes Receptor Internalization
The ability of 1-15D1 to promote the T-cell response was evaluated, since Siglec-15 is an effective immunosuppressive molecule. Human PBMCs from a healthy donor were stimulated with an anti-CD3 mAb (OKT3) plus Siglec-15 protein and 1-15D1 for 72 hours. Cell viability was assessed with CellTiter-Glo reagent (Fig. 3A). The IFN-γ secretion levels was measured after 24 hours in the absence or presence of antibodies. The results showed that 1-15D1 effectively abrogated the inhibitory effect of T cells, and a profound increase in the level of secreted IFN-γ was observed compared with that induced by 5G12 (Fig. 3B). Furthermore, we explored the relationship between drug dose and drug effect on IFN-γ secretion after treatment with 1-15D1. The dose-response curve of IFN-γ secretion revealed graded responses of 1-15D1, with an EC50 of 26.19 μg/mL (Fig. 3C).
FIGURE 3: Activation of human PBMCs and antibody-mediated receptor internalization. A, PBMCs were treated with 50 μg/mL 1-15D1 or 5G12 for 72 hours and cell proliferation was measured using CellTiter-GLO. hIgG1 was used as a negative control. B and C, PBMCs were treated with 50 μg/mL (B) or serially diluted (from 3.13 to 100 μg/mL) (C) 1-15D1 or 5G12 for 24 hours, and the level of secreted IFN-γ was measured by ELISA assay. hIgG1 was used as a negative control. D, Hek293 (control) and Hek293-hsiglec-15 cells were incubated with 1-15D1 at 37°C or 4°C for 20 hours. Total fluorescence intensities were measured by FACS. The histogram shows Siglec-15 receptor internalization following treatment with 1-15D1 (5 μg/mL) at 37°C (deepgray), 4°C (lightgray) or hIgG (gray). E, Hek293 and Hek293-hsiglec-15 cells were incubated for 24 hours with pHAb-labeled antibodies. Representative images show punctate internalized 1-15D1 in Hek293-hsiglec-15 cells. Cells were analyzed by fluorescence microscopy. Scale bar, 50 μm.
Furthermore, we investigated whether 1-15D1 exacerbated Siglec-15 internalization. Hek293-hsiglec-15 cells were incubated with 1-15D1 for 20 hours at 37°C, and the antibody signal on the cell surface was decreased (Fig. 3D). This phenomenon was not observed when cells were incubated with 1-15D1 at 4°C. The decrease in total signal may have been due to the internalization of the surface receptor. Therefore, we employed a fluorescence-based internalization assay using mAb labeled with pHAb. Robust fluorescence was observed in Hek293-hsiglec-15 cells but not in Hek293 cells after incubation with pHAb-labeled 1-15D1 for 20 hours (Fig. 3E). These results indicated that 1-15D1 recognizes Siglec-15 on the cell surface and promotes Siglec-15 internalization.
Anti-Siglec-15 mIgG2a is More Effective Than Anti-Siglec-15 mIgG1 and mIgG2aDANA in a Tumor Model
A humanized mouse, which was engineered to express certain human proteins,35 is a powerful tool in the in vivo study of human cancer immunology and immunotherapy. A schematic diagram of mouse model establishment and the PCR strategy used for transgenic founder mouse screening is shown in Figure 4A. To assess the effect of the anti-Siglec-15 isotype on the antitumor effects in vivo, 2 isotype mAbs, 1-15D1-mIgG2a and 1-15D1-mIgG1 were generated. MC38 cells overexpressing human Siglec-15 (MC38-hS15 cells) were implanted subcutaneously into homozygous Siglec-15 humanized mice. After the tumor volume reached ~80Â mm3, the mice were treated (biw for 3Â wk at 15Â mg/kg) with 1-15D1-mIgG2a and 1-15D1-mIgG1. On day 17, the average tumor size of the vehicle treatment group reached 1725Â mm3, whereas 1-15D1-mIgG2a and 1-15D1-mIgG1 treatment groups showed a 57% and 32% reduction compared with the untreated mice, respectively (Fig. 4B).
Figure 4: Humanized mice for the study of anti-cancer effects in vivo. A, Schematic diagram of the transgene structure and PCR strategy of transgenic founder mouse screening. The Siglec-15 transgene mice harbored partial human Siglec-15 sequences of exon 2, exon 3, and part of exon 4, as shown in dark gray. Primers for the 5’ arm and 3’ arm are indicated. Two 3.7 kb fragments were amplified by PCR in transgenic mice. Lane 1: wild-type littermate; Lanes 2–3: transgenic mice. B–E, Mice were implanted with MC38-hS15 cells. When tumors were ~80 mm3 in volume, they were treated with the following: 1-15D1-mIgG1, 1-15D1-mIgG2a, 1-15D1-mIgG2a DANA. Treatment was given twice weekly for 17 or 13 days, starting on day 6 postimplantation. Saline was used as a vehicle control (n=4 per group). Tumor volume (B, C), and body weight (D, E) were measured. (F) We immobilized mFcγRI, mFcγRIIb, mFcγRIII, and mFcγRIV on sensors and assessed the binding of our isotypes to the different receptors via BLI.
Because a difference in the efficacy of the mIgG1 and mIgG2a anti-Siglec-15 antibodies was found in the humanized mouse model, we wanted to determine whether the effect can be attributed to an ADCC or ADCP mechanism. 1-15D1-mIgG2a with FcγR-mediated effector function-null mutations D265A and N297A (DANA) were generated and tested. The result showed that 1-15D1-mIgG2a-DANA exerted no significant antitumor effect compared with that of 1-15D1-mIgG2a (Fig. 4C). No body weight changes were observed with the indicated treatments (Figs. 4D, E).
Next, the binding affinities of 1-15D1-mIgG2a, 1-15D1-mIgG2aDANA, and 1-15D1-mIgG1 for murine FcγRs were measured via BLI. In summary, the mIgG2a isotype showed similar mFcγR-binding affinities, but the mIgG2aDANA variant did not bind all the mFcγRs, while mIgG1 did not bind mFcγRI or mFcγRIV but bound mFcγRIIb, in accordance with previous reports36,37 (Fig. 4F). Moreover, all mAbs had high purity and exhibited a similar-binding affinity for hsiglec-15 (Supplemental Digital Content Fig. 1, Supplemental Digital Content 3, https://links.lww.com/JIT/A716). As previously reported, the NK cells of mice express only mFcγRIII, and macrophages variably express all FcγRs.38 Collectively, our studies demonstrated that the 1-15D1-mIgG2a isotype led to high antitumor efficacy due to both ADCC and ADCP mechanisms.
Epitope Binning of mAbs
BLI was used to identify the epitope bins.39 In this assay, we performed sought to determine whether the mAbs compete for the same binding site on Siglec-15. Sensors were loaded with recombinant Siglec-15. Subsequently, the sensors were exposed to the first mAb, and after binding equilibration, the sensors were exposed to a second antibody (Fig. 5A). Simultaneous binding induces a wavelength shift, which indicates that the 2 antibodies bind to independent epitopes. In contrast, when the wavelength does not shift or shifts very little, two antibodies have bound the same or partially overlapping epitope. The results suggested that 1-15D1 did not compete with 5G12, indicating that they bound distinct sites on Siglec-15 (Fig. 5B). These results were graphically depicted in Figure 5C, which indicated that we screened a new recognition epitope mAb.
Figure 5: Epitope binning of monoclonal antibodies (mAbs). A, The cartoon showed the classical sandwich assay format used. The antigen was immobilized on the biosensor, followed by the binding of the saturating mAb (first mAb) and competing mAb (second mAb). B, Epitope binning analysis of 2 antibodies using Octet. C, Summarizing the epitope binning results, where gray boxes indicate noncompeting antibody pairs, and black boxes indicate self-competition.
DISCUSSION
Recent studies have highlighted the important role of Siglec-15 in modulating the tumor microenvironment and promoting tumor suppression.21 In addition to TAMs, some tumor cells upregulate Siglec-15 expression, which contributes to an immunosuppressive microenvironment by blocking CD8+ T-cell proliferation.20,21 In our study, 1-15D1 showed a clear ability to reverse Siglec15-mediated inhibition of T-cell activity. Importantly, we investigated the Fc-mediated effector functions of 1-15D1 in mouse models of colorectal cancer by evaluating different IgG subclasses. To the best of our knowledge, this is the first study to focus on the Fc effector function of an anti-Siglec15 mAb.
As previously reported, the ligand of Siglec-15 is very extensive, and many ligands can bind Siglec-15.34,40,41 Some ligands are overexpressed on the surface of many tumor cells, such as Siglec-15 on M2-like macrophages, which has been found to interact with Sialyl-Tn carbohydrate on tumor cells.9,20,28,42 There is evidence suggesting that substitution of Arg at position 143 eliminates sialic acid binding and the loss of the immune regulatory function of Siglec-15.8,9 It is difficult to prepare high-affinity mAbs, especially for binding specific epitopes. In the present study, based on a Siglec-15 molecular signal transduction mechanism, we initially screened a novel epitope mAb that recognized a different epitope with 5G12 and interfered with the interaction between LRRC4C and Siglec-15 (Figs. 1, 2, 5). More importantly, this mAb showed many excellent properties, such as high affinity for Siglec-15 and significant recovery of Siglec-15-mediated suppression of T cells in vitro compared with 5G12 (Figs. 1, 3).
Endocytosis is important for the therapeutic effect and often depends on the target antigen and the epitope recognized by an antibody.43,44 Since 1-15D1 showed high potency in reversing Siglec-15-mediated suppression of human T cells in vitro, we sought to reveal the cellular mechanisms underlying 1-15D1 action. Thus, we found that 1-15D1 induced Siglec-15 receptor internalization in Hek293-hsiglec-15 cells, which may negatively regulated Siglec-15 signaling (Fig. 3). This phenomenon was consistent with the investigation of Matthew Stuible and colleagues that a Siglec-15 mAb induced receptor internalization and degradation in osteoclasts.12 Collectively, these results demonstrated that dual mechanisms of action were induced by 1-15D1 for the therapeutic effect in vitro, through blocking Siglec-15 interaction with a receptor on T cells and inducing the internalization of cell surface Siglec-15.
To date, the antitumor efficacy of Siglec-15-specific mAbs in mouse tumor models has rarely been reported.21,29 In our study, 1-15D1 bound to human and cynomolgus Siglec-15 with similar affinity but did not bind to mouse Siglec-15. Whereas, we used a human Siglec-15 knock-in mouse, which expressed the human extracellular domain of Siglec-15, as a preclinical model for the in vivo study. Notably, this is the first study to establish an animal xenograft model using human Siglec-15 knock-in mice. In addition, it had been reported that functional Fc contributes to therapeutic efficacy in mouse models, such as anti-CTLA-4 or Tigit checkpoint inhibitors.30,32,45 Considering that the selection of the correct Ig isotype is crucial, we examined the isotype-dependent efficacy of the antitumor response induced by 1-15D1. We generated 2 different isotype mAbs using mouse IgG2a, IgG1, and the FcγR binding-deficient variant mIgG2a-DANA, which carried the same variable region, to examine the relevance of Fc-mediated therapeutic outcomes of 1-15D1 treatment in mouse models of colorectal cancer. Treatment with the mIgG2a antibody resulted in pronounced antitumor activity in vivo, whereas that with the mIgG2a-DANA and mIgG1 antibodies showed minimal or no antitumor activity. To better understand why mIgG2a was more effective, we assessed the binding affinities of mIgG2a, mIgG2aDANA, and mIgG1 for murine FcγRs. As expected, mIgG2aDANA did not bind all the mFcγRs (Fig. 4). The results suggested that enhanced ADCC/ADCP resulted in enhanced antitumor activity and survival. However, our experiments also had some limitations. Although we found significant effects in the mIgG2a treatment groups, the precise Fc-mediated mechanisms involved remain under investigation.
There are no marketed drugs targeting Siglec-15, and rare clinical projects are currently underway, it is an urgent need to provide patients with more Siglec-15 antibody drugs with better therapeutic effects. Normalization cancer immunotherapy is the future trend for the eradication of human cancers. Our results suggested that 1-15D1, with the scientific rationale for clinical development and Fc-mediated effector functions may contribute to the optimal response to the activity of anti-Siglec-15 mAbs.
CONFLICTS OF INTEREST/FINANCIAL DISCLOSURES
None reported. All authors have declared that there are no financial conflicts of interest with regard to this work.
REFERENCES
1. Iwai Y, Hamanishi J, Chamoto K, et al. Cancer immunotherapies targeting the PD-1 signaling pathway. J Biomed Sci. 2017;24:26.
2. Doi T, Piha-Paul SA, Jalal SI, et al. Safety and antitumor activity of the anti-programmed death-1 antibody pembrolizumab in patients with advanced esophageal carcinoma. J Clin Oncol. 2018;36:61–67.
3. Haslam A, Prasad V. Estimation of the percentage of US patients with cancer who are eligible for and respond to checkpoint inhibitor immunotherapy drugs. JAMA Netw Open. 2019;2:e192535.
4. McDermott D, Haanen J, Chen TT, et al. Efficacy and safety of ipilimumab in metastatic melanoma patients surviving more than 2 years following treatment in a phase III trial (MDX010-20). Ann Oncol. 2013;24:2694–2698.
5. Gettinger SN, Horn L, Gandhi L, et al. Overall survival and long-term safety of nivolumab (anti-programmed death 1 antibody, BMS-936558, ONO-4538) in patients with previously treated advanced non-small-cell lung cancer. J Clin Oncol. 2015;33:2004–2012.
6. Economopoulou P, Perisanidis C, Giotakis EI, et al. The emerging role of immunotherapy in head and neck squamous cell carcinoma (HNSCC): anti-tumor immunity and clinical applications. Ann Transl Med. 2016;4:173.
7. Stanczak MA, Siddiqui SS, Trefny MP, et al. Self-associated molecular patterns mediate cancer immune evasion by engaging Siglecs on T cells. J Clin Invest. 2018;128:4912–4923.
8. Angata T, Tabuchi Y, Nakamura K, et al.
Siglec-15: an immune system Siglec conserved throughout vertebrate evolution. Glycobiology. 2007;17:838–846.
9. Murugesan G, Correia VG, Palma AS, et al.
Siglec-15 recognition of sialoglycans on tumor cell lines can occur independently of sialyl Tn antigen expression. Glycobiology. 2021;31:44–54.
10. Hiruma Y, Hirai T, Tsuda E.
Siglec-15, a member of the sialic acid-binding lectin, is a novel regulator for osteoclast differentiation. Biochem Biophys Res Commun. 2011;409:424–429.
11. Hiruma Y, Tsuda E, Maeda N, et al. Impaired osteoclast differentiation and function and mild osteopetrosis development in
Siglec-15-deficient mice. Bone. 2013;53:87–93.
12. Stuible M, Moraitis A, Fortin A, et al. Mechanism and function of monoclonal antibodies targeting
Siglec-15 for therapeutic inhibition of osteoclastic bone resorption. J Biol Chem. 2014;289:6498–6512.
13. Shimizu T, Takahata M, Kameda Y, et al. Sialic acid-binding immunoglobulin-like lectin 15 (
Siglec-15) mediates periarticular bone loss, but not joint destruction, in murine antigen-induced arthritis. Bone. 2015;79:65–70.
14. von Gunten S, Bochner BS. Basic and clinical immunology of Siglecs. Ann N Y Acad Sci. 2008;1143:61–82.
15. Kameda Y, Takahata M, Komatsu M, et al.
Siglec-15 regulates osteoclast differentiation by modulating RANKL-induced phosphatidylinositol 3-kinase/Akt and Erk pathways in association with signaling Adaptor DAP12. J Bone Miner Res. 2013;28:2463–2475.
16. Ishida-Kitagawa N, Tanaka K, Bao X, et al.
Siglec-15 protein regulates formation of functional osteoclasts in concert with DNAX-activating protein of 12 kDa (DAP12. J Biol Chem. 2012;287:17493–17502.
17. Angata T.
Siglec-15: a potential regulator of osteoporosis, cancer, and infectious diseases. J Biomed Sci. 2020;27:10.
18. Duque G, Huang DC, Dion N, et al. Interferon-gamma plays a role in bone formation in vivo and rescues osteoporosis in ovariectomized mice. J Bone Miner Res. 2011;26:1472–1483.
19. Kameda Y, Takahata M, Mikuni S, et al.
Siglec-15 is a potential therapeutic target for postmenopausal osteoporosis. Bone. 2015;71:217–226.
20. Takamiya R, Ohtsubo K, Takamatsu S, et al. The interaction between
Siglec-15 and tumor-associated sialyl-Tn antigen enhances TGF-beta secretion from monocytes/macrophages through the DAP12-Syk pathway. Glycobiology. 2013;23:178–187.
21. Wang J, Sun J, Liu LN, et al.
Siglec-15 as an immune suppressor and potential target for normalization
cancer immunotherapy. Nat Med. 2019;25:656–666.
22. Cao X, Zhou Y, Mao F, et al. Identification and characterization of three Siglec15-related immune and prognostic subtypes of breast-invasive cancer. Int Immunopharmacol. 2022;106:108561.
23. Liang H, Chen Q, Hu Z, et al. Siglec15 facilitates the progression of non-small cell lung cancer and is correlated with spinal metastasis. Ann Transl Med. 2022;10:281.
24. Liu W, Ji Z, Wu B, et al.
Siglec-15 promotes the migration of liver cancer cells by repressing lysosomal degradation of CD44. FEBS Lett. 2021;595:2290–2302.
25. Yang WB, Qin CP, Du YQ, et al.
Siglec-15 promotes progression of clear renal cell carcinoma. Chin Med J (Engl). 2021;134:2635–2637.
26. Li B, Zhang B, Wang X, et al. Expression signature, prognosis value, and immune characteristics of
Siglec-15 identified by pan-cancer analysis. Oncoimmunology. 2020;9:1807291.
27. Rodrigues Mantuano N, Natoli M, Zippelius A, et al. Tumor-associated carbohydrates and immunomodulatory lectins as targets for
cancer immunotherapy. J Immunother Cancer. 2020;8:e001222.
28. Julien S, Videira PA, Delannoy P. Sialyl-tn in cancer: (how) did we miss the target? Biomolecules. 2012;2:435–466.
29. He F, Wang N, Li J, et al. High affinity
monoclonal antibody targeting
Siglec-15 for
cancer immunotherapy. J Clin Transl Res. 2021;7:739–749.
30. Yang F, Zhao L, Wei Z, et al. A cross-species reactive TIGIT-blocking antibody Fc dependently confers potent antitumor effects. J Immunol. 2020;205:2156–2168.
31. Arce Vargas F, Furness AJS, Litchfield K, et al. Fc effector function contributes to the activity of human anti-CTLA-4 antibodies. Cancer Cell. 2018;33:649–63 e4.
32. Dahan R, Sega E, Engelhardt J, et al. FcgammaRs modulate the anti-tumor activity of antibodies targeting the PD-1/PD-L1 axis. Cancer Cell. 2015;28:543.
33. Bornhofft KF, Martorell Ribera J, Viergutz T, et al. Characterization of Sialic acid-binding immunoglobulin-type lectins in fish reveals teleost-specific structures and expression patterns. Cells. 2020;9:836.
34. Sun J, Lu Q, Sanmamed MF, et al. Correction:
Siglec-15 as an emerging target for next-generation
cancer immunotherapy. Clin Cancer Res. 2021;27:3804.
35. Frese KK, Tuveson DA. Maximizing mouse cancer models. Nat Rev Cancer. 2007;7:645–658.
36. Tipton TR, Roghanian A, Oldham RJ, et al. Antigenic modulation limits the effector cell mechanisms employed by type I anti-CD20 monoclonal antibodies. Blood. 2015;125:1901–1909.
37. Nimmerjahn F, Bruhns P, Horiuchi K, et al. FcgammaRIV: a novel FcR with distinct IgG subclass specificity. Immunity. 2005;23:41–51.
38. Jonsson F, Mancardi DA, Albanesi M, et al. Neutrophils in local and systemic antibody-dependent inflammatory and anaphylactic reactions. J Leukoc Biol. 2013;94:643–656.
39. Abdiche YN, Malashock DS, Pinkerton A, et al. Exploring blocking assays using Octet, ProteOn, and Biacore biosensors. Anal Biochem. 2009;386:172–180.
40. Chang L, Chen YJ, Fan CY, et al. Identification of Siglec ligands using a proximity labeling method. J Proteome Res. 2017;16:3929–3941.
41. Briard JG, Jiang H, Moremen KW, et al. Cell-based glycan arrays for probing glycan-glycan binding protein interactions. Nat Commun. 2018;9 880.
42. Munkley J. The Role of Sialyl-Tn in Cancer. Int J Mol Sci. 2016;17:275.
43. Hurwitz E, Stancovski I, Sela M, et al. Suppression and promotion of tumor growth by monoclonal antibodies to ErbB-2 differentially correlate with cellular uptake. Proc Natl Acad Sci U S A. 1995;92:3353–3357.
44. Yarden Y. Agonistic antibodies stimulate the kinase encoded by the neu protooncogene in living cells but the oncogenic mutant is constitutively active. Proc Natl Acad Sci U S A. 1990;87:2569–2573.
45. Simpson TR, Li F, Montalvo-Ortiz W, et al. Fc-dependent depletion of tumor-infiltrating regulatory T cells co-defines the efficacy of anti-CTLA-4 therapy against melanoma. J Exp Med. 2013;210:1695–1710.
