Biomedicines | Free Full-Text | T Cell Receptor-Directed Bispecific T Cell Engager Targeting MHC-Linked NY-ESO-1 for Tumor Immunotherapy

By inergency On Apr 1, 2024

[ad_1]

1. Introduction

The T cell engager (TCE), which redirects cytotoxic T cells to kill tumor cells, is emerging as a promising strategy to treat cancer patients, especially those with a low-mutation-burden tumor or a low number of tumor-specific T cells who do not respond to checkpoint inhibitors [1]. In general, the TCE is comprised of two domains: one targets the tumor antigen on the tumor cell surface and the other binds the protein (typically the CD3 delta) on the T cell membrane. Over decades, there has been much effort directed toward the design of TCE formats based on the variable domain of antibodies. To date, more than 20 platforms have made their way into clinical development [2], and eight TCEs have been approved by the FDA. Although the efficacy of these drugs is impressive, most TCEs cause harmful ‘on-target off-tumor’ effects due to the unfavorable specificity of tumor-associated antigen (also expressed on normal cells) on the tumor cells [3]. On the other hand, it is challenging to seek tumor-specific antigens, as the target antigens with an extracellular domain that is accessible to antibodies comprise only 4], and the mutations that differentiate tumor cells from normal cells may be buried inside of the protein.

Major histocompatibility complex (MHC) can present peptides derived from intracellular proteins (>90% of the proteome) to the cell surface to form peptide–MHC (pMHC), which is recognized by the T cell receptor (TCR) on T cells. The pMHCs extend the range of antigens on the membrane and are regarded as tumor-specific antigens when the peptide is tumor-specific (usually mutated or with highly restricted expression in normal tissues) [5]. However, employing TCR as a soluble drug to target pMHC is challenging due to the poor water solubility of TCR [6]. Over decades, several protein-engineering solutions have been developed to generate stable and soluble TCRs, including stabilizing mutations of the TCR C domain, fusing a soluble protein part to the TCR C terminus, and other bespoke stabilizations of individual TCRs [6]. The first reported TCR-based TCE was ImmTACm which comprises an affinity-matured TCR fused to a humanized CD3-specific single-chain antibody fragment (scFv) [7] and exhibits high efficiency in killing tumor cells (EC₅₀ at picomole level). One ImmTAC molecule named Tebentafusp has been approved by the FDA for treating uveal melanoma [8]. However, the half-life of ImmTAC is only several hours, which requires more frequent dosing [7], and it is not straightforward to produce ImmTAC with the correct structure due to the complex refolding procedure. Therefore, more formats of T-TCE with distinct properties need to be explored.

The crystalline fragment (Fc) of an antibody is a widely used part for fusing to improve the half-life and water solubility of a protein. Recently, soluble expression of T-TCE with Fc in mammalian cells was reported by Froning et al. with a simple method [9]. Inspired by this, our team previously reported a soluble click-linked T-TCE but the killing efficiency of T-TCE in single uses was not satisfactory [10]. Herein, we describe a new IgG-like T-TCE (IgG-T-TCE) format with the tumor antigen-targeting arm (VH and VL) of IgG replaced by an evolved TCR. The IgG-T-TCE molecules were expressed in mammalian cells following simple purification (Ni column and anti-Flag column). For evaluating the format, we set a well-studied paired TCR/pMHC as a representative: 1G4-113 TCR and NY-ESO-1_157–165/HLA-A*02:01 (NY-ESO-1, a cancer–testis antigen; HLA*A-02:01, one of the most common MHC alleles in human) [11]. The purified product showed retained affinities to the tumor target and human CD3 and the ability to activate T cells and specifically kill target tumor cells in vitro. We also evaluated the impact of various anti-CD3 antibodies and Fc silencing on the in vitro activities of the IgG-T-TCE-NY. Finally, the IgG-T-TCE-NY efficiently inhibited tumor growth in a tumor–PBMC co-engrafted mouse model without any obvious toxicities, demonstrating its potential for precise tumor immunotherapy in the future.

2. Materials and Methods

2.1. Cells and Culture Conditions

Human embryonic kidney 293F cells (HEK293F) were kindly provided by the Comprehensive AIDS Research Center (Tsinghua University, Beijing, China) and rotary-cultured in an SMM 293-TI medium (Sino Biological Inc., Beijing, China).

Cell lines A375 (melanoma), Jurkat (T cell leukemia), and K562 (chronic myelogenous leukemia) were purchased from American Type Culture Collection (ATCC, Manassas, VA, USA). Cell line U266 (multiple myeloma) was purchased from BeNa Culture Collection (BNCC, Beijing, China).

The construction of the A375-NY-GFP, K562-NY (NY-ESO-1_157–165⁺, HLA-A*02:01⁺, GFP⁺) and K562-Ctrl (irrelevant peptide, HLA-A*02:01⁺, GFP⁺) cell lines has been described previously [10].

All tumor cells here were maintained in RPMI-1640 (Gibco, Grand Island, NY, USA) with 10% fetal bovine serum (FBS; Gibco, Grand Island, NY, USA) at 37 °C with 5% CO₂ and 95% humidity.

Human peripheral blood mononuclear cells (PBMCs) were purchased from Shanghai Saily Biotechnology Co., Ltd. (Shanghai, China). Generally, PBMCs were allowed to rest in RPMI-1640 with 10% FBS for several hours before use.

2.2. Design, Construction, Expression, and Purification of IgG-T-TCE-NY

The design of the IgG-like T cell engager was based on the backbone of conventional antibody IgG1 (Figure 1a). Briefly, two binding domains targeting pMHC and human CD3 were located at two short arms in a TCR-fused format and a Fab, respectively. The TCR consisted of the variable domain and partial constant domain that was extracellular. The constant domains of the TCR and antibody were linked by a GS linker and a pair of cysteine mutations (T48C in TRAC, S57C in TRBC2) was inserted to improve the stability of the TCR. For reducing mis-paired products, the ‘knob into hole’ strategy for Fc and a CrossMAb design were applied. Moreover, two light chains were fused with the His tag and Flag tag, respectively, to improve the purity of the purified products. 1G4-113 TCR targeting NY-ESO-1_157–165/HLA-A*02:01 [12] was set as a representative TCR and a humanized OKT3 clone was chosen to bind human CD3. The genes encoding four chains of IgG-T-TCE-NY were cloned into vector pLVX-Puro and transiently transfected into HEK293F cells using linear polyethyleneimine (PEI; Polysciences, Warrington, PA, USA) together.

The supernatant of HEK293F was collected after 3–4 days’ culture and filtered through a 0.45 μm filter unit, followed by two-step purification using a Ni column (HisTrapTM HP; GE Healthcare Life Sciences, Little Chalfont, Buckinghamshire, UK) and Anti-Flag M2 Affinity Gel (Millipore Sigma, St. Louis, MO, USA). The purified product was then concentrated in phosphate buffered saline (PBS), aliquoted, and kept at −80 °C for long-term storage.

As for the variants of IgG-T-TCE-NY, Fc was silenced by LALAPG substitutions and OKT3 was replaced with OKT3-LT (lower affinity to CD3 [13]) or UCHT1 (higher affinity to CD3 [14]).

2.3. Characterization of IgG-T-TCE-NY by SDS-PAGE, WB analysis, ELISA, and Flow Cytometry

SDS-PAGE and WB analysis were performed to identify the components of the purified product with a common operation. As a note, deglycosylation was conducted using the Enzymatic In-Solution N-Deglycosylation Kit (Sigma, St. Louis, MO, USA) overnight before reducing SDS-PAGE; the chains with a His tag or Flag tag were labeled with His Tag (C-terminal Specific) Mouse Monoclonal Antibody (Beyotime, Shanghai, China) or Flag Tag Mouse Monoclonal Antibody (Beyotime, Shanghai, China).

The affinity of IgG-T-TCE-NY to pMHC was tested via ELISA as previously described [10]. Briefly, a 96-well EIA/RIA plate was coated with streptavidin (2 µg/mL) overnight at 4 °C. After blocking, biotin-labeled pMHCs (2 µg/mL, generated by refolding as in a previous report [15]) were added for 2 h at 37 °C. Then, IgG-T-TCE-NY and its controls were added for 1 h at 37 °C, followed by incubation with HRP-conjugated goat anti-human IgG (H + L) (1:1000 dilution, Beyotime, Shanghai, China) for 1 h at 37 °C. The samples finally reacted with the TMB substrate solution and the reaction was stopped by adding 2M H₂SO₄. The absorbance of samples at 450 nm was measured using a Model 680 Microplate Reader (Bio-Rad, Hercules, CA, USA). Washing with PBST was needed between each step.

The affinity of IgG-T-TCE-NY to human CD3 was tested via flow cytometry as previously described [10]. Jurkat cells (expressing human CD3) were used as target cells. Briefly, target cells were harvested and incubated with serial dilutions of IgG-T-TCE-NY or controls for 30 min at 4 °C. Then, cells were incubated with FITC-conjugated goat anti-human IgG (H + L) (1:200 dilution, Beyotime, Shanghai, China) for 30 min at 4 °C. Samples were analyzed with an ACEA NovoCyte^TM flow cytometer (ACEA Biosciences, San Diego, CA, USA). Washing with PBS was needed between each step. In the experiment testing simultaneous binding of IgG-T-TCE-NY to NY-ESO-1_157–165/HLA-A*02:01 and human CD3, similar operations were executed, except incubation with a biotin-labeled pMHC and SA-PE was used instead of FITC-conjugated goat anti-human IgG following the IgG-T-TCE-NY incubation.

2.4. In Vitro Cytotoxicity Assay

The cytotoxicity of tri-specific T cell engagers was evaluated using apoptosis detection or the LDH-releasing assay. Human PBMCs and target tumor cells were incubated in a 4:1 ratio in the presence of serial dilutions of tri-specific T cell engagers or controls in both experiments. For apoptosis (constructed cell lines that carried GFP), mixed cells were generally incubated in 48-well plates in RPMI-1640 with 10% FBS. After 2 days, samples were harvested, stained using an Annexin V 633 Apoptosis Detection Kit (Dojindo, Kumamoto, Japan), and analyzed with an ACEA NovoCyte^TM flow cytometer. For LDH, mixed cells were generally incubated in 96-well plates in RPMI-1640 (without phenol red) with 1.5% FBS. Additional control wells were set up with PBMCs alone, tumor cells alone, or medium alone to calculate the spontaneous LDH value from cells. After 2–3 days, the supernatants were transferred into new 96-well plates. Following the steps of the LDH Cytotoxicity Assay Kit (Beyotime, Shanghai, China), the released LDH was represented by the absorbance at 490 nm. The lysis percentage was calculated as (experimental well − spontaneous PBMCs − spontaneous tumor cells − medium alone)/(MAX tumor cells alone − tumor cells alone) × 100. Three repeats were used for each sample.

2.5. T Cell Activation, Proliferation, and Cytokine Analysis

The early activation marker CD69 and late activation marker CD25 on the surface were evaluated using flow cytometry to confirm the activation state of T cells. Briefly, cells were harvested after 2 days of co-culture and stained with the following antibodies before detection: Super Bright 436 anti-human CD69 (Invitrogen, Carlsbad, CA, USA), PE anti-human CD25 (Invitrogen, Carlsbad, CA, USA), and APC anti-human CD3 (Invitrogen), APC anti-human CD8 (Invitrogen, Carlsbad, CA, USA), or APC anti-human CD4 (Invitrogen, Carlsbad, CA, USA).

For T cell proliferation, PBMCs were labeled with CFSE (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s protocol before co-culture. After 3–8 days of co-culture, cells were harvested, stained with PI (Invitrogen, Carlsbad, CA, USA), and analyzed with flow cytometry.

Cytokine IL-2 and Granzyme B were detected using ELISA and flow cytometry, respectively. Cells and the supernatants were collected after 2–3 days of co-culture. IL-2 in the supernatant was measured according to the Human IL-2 Precoated ELISA kit (Dakewe, Shenzhen, China). As for intracellular Granzyme B, cells were fixed and permeabilized using an eBioscience™ Intracellular Fixation and Permeabilization Buffer Set (Invitrogen, Carlsbad, CA, USA) as indicated by the manufacturer before detection.

2.6. In Vivo Antitumor Activity

Fifteen eight-week-old female NOD-SCID mice (GemPharmatech, Nanjing, China) were randomly divided into three groups and given different treatments: no PBMC group—A375 (s.c.) and PBS (i.v.); vehicle control group—PBMC + A375 (s.c.) and PBS (i.v.); IgG-T-TCE group—PBMC + A375 (s.c.) and IgG-T-TCE-NY (i.v.). In detail, mice were subcutaneously engrafted with 1.5 × 10⁶ tumor cells and 3.75 × 10⁶ PBMCs together or tumor cells alone in the right armpit on day 0. Each group was administrated intravenously with 105 µg/kg IgG-T-TCE-NY or PBS on day 0, day 3, and day 6. The tumor volume was monitored and recorded according to the following formula: volume = (length × width²)/2. Mice were euthanized when there was a mouse with a tumor larger than 800 mm³ and tumors were taken from mice. The animal work was performed following the protocol approved by the Committee on the Ethics of Animal Experiments of Zhejiang University (Hangzhou, China, 17526).

4. Discussion

In this study, we developed an easy-to-obtain TCR-based TCE in an IgG form (IgG-T-TCE) that redirected T cells to specifically and efficiently kill tumor cells with target peptide–MHCs in vitro and to control tumor growth in vivo.

Simply put, the constant (extracellular part) and variable regions of TCR were connected to the constant region of the antibody in one arm, and a pair of cysteines was inserted in the constant region of the TCR to stabilize the TCR. The IgG-T-TCE was obtained through two-step purification (His tag and Flag tag) from the eukaryotic expression system and showed as simultaneous binding targets pMHC and hCD3. Differently from prokaryotic sources, the TCR part in our IgG-T-TCE obtained from the eukaryotic expression system was obviously glycosylated. This N-glycosylation was predicted to be due to multiple glycosylated sites on TCR. It has been clearly shown that glycosylation of TCR on T cells can protect TCR from proteases, increase steric hindrance to decrease nonspecific aggregation, and confer rigidity to TCR to increase interaction with pMHC [16,17]. However, one study showed that T cells with deglycosylated TCR had enhanced avidity to tumor cells and the pMHC tetramer [18], implying that glycosylation may restrict TCR’s binding to pMHC. This needs to be further confirmed because increased TCR clustering could also explain it. We simply compared the binding of our IgG-T-TCE with the pMHC monomer (prokaryotic sources) before and after N-deglycosylation via ELISA. Compared to the ‘before’ group, the ‘after’ group showed no advantage in binding affinity (and even a slight decrease at low concentration) (Figure 1f). One question is whether the glycans on pMHC interact. N-glycosylation inhibition experiments showed that the absence of N-glycans does not modify pMHC’s ability to interact with TCR [16].

The specific lysis of tumor cells mainly depends on the specificity of molecules towards targets. For pMHC, which has a low density on the cell surface (e.g., for NY-ESO-1_157–165/HLA-A*02:01, 10–50 copies per cell [19]), a monovalent molecule with a high affinity is more effective in enriching molecules on the tumor cell side than a bivalent molecule with a low affinity, which requires a mature screening platform to obtain a specific clone with a high affinity. Compared with the TCR platform [20,21,22], the antibody platform is relatively easy to establish. Some researchers have attempted to develop TCR mimic antibodies to target pMHC [23], but most TCR mimic antibodies exhibited a greater degree of cross-reactivity due to the different recognition modes of TCRs and antibodies towards pMHC [24], which was not conducive to overall specificity. Here, we used a well-identified pMHC (NY-ESO-1_157–165/HLA-A*02:01) as the cancer-specific antigen and selected a highly specific TCR clone with a very high affinity (1G4-113 clone, Kd at picomole level [12]) to target it. To evaluate the specificity of this IgG-T-TCE, K562-NY (transfected with a plasmid expressing NY-ESO-1_157–165) and K562-Ctrl (transfected with a plasmid expressing irrelevant peptide) were established as a positive cell line and negative cell line, respectively, based on the same parent cell line K562-A2 we established previously. We compared the apoptosis of two cell lines in the same co-culture condition (PBMC + drug + tumor cell), and the result showed that positive cells demonstrated much more apoptosis than negative cells, indicating this IgG-T-TCE targeted NY-ESO-1_157–165/HLA-A*02:01 but did not cross-react with the irrelevant one tested here.

We also evaluated the killing efficiency of IgG-T-TCE-NY on cell line A375. The EC₅₀ of IgG-T-TCE-NY here was at the picomole level and influenced by multiple factors: with PBMCs, a certain affinity to CD3 was needed for IgG-T-TCE to sufficiently kill tumor cells but with a threshold (Figure 3e); and silence of Fc (IgG1) weakened the ability of IgG-T-TCE (Figure 3d), implying the engagement of FcγR receptor positive immune cells in the elimination of tumor cells. In TCE, Fc acts as a ‘rapier’, adding additional mechanisms of cytotoxicity, like phagocytosis or antibody-dependent cell-mediated cytotoxicity, but also increasing non-specific activation of T cells, leading to a higher risk of cytokine storms [25]. A representative of TCEs that reserve IgG function is Catumaxomab, which was approved in 2009 by the EMA for the treatment of malignant ascites but with a careful administration method for acceptable safety [26]. Besides the dose interruption, additional molecules like kinase inhibitors or cytokine antagonists may reduce cytokine release [27,28,29].

TCEs have shown some clinical efficacy [30]. However, activating T cell receptor signaling with anti-CD3 would not stimulate T cells in a fully activated state, which needs co-stimulatory signaling (CD28, 4-1BB, OX40, etc.) [31]. Recently, a trispecific antibody that interacts with CD38 on tumor cells and CD3 and CD28 on T cells displayed enhanced cytotoxicity to tumor cells [32]. Similar results can be seen for another trispecific antibody that interacts with HER2, CD3, and CD28 [33]. These studies showed that multi-specific TCEs with the synergy from multiple mechanisms may be the next-generation T cell immunotherapy. Benefiting from the good extension ability of IgG, IgG-T-TCE has the potential to be developed into different multi-specific TCEs with high specificity to tumor cells, which is worth exploring.

[ad_2]