https://orcid.org/0000-0002-4516-8355
ABSTRACT
The clinical successes of immune checkpoint blockade have invigorated efforts to activate T cell-mediated responses against cancer. Targeting members of the PVR family, consisting of inhibitory receptors TIGIT, CD96, and CD112R, has been an active area of clinical investigation. In this study, the binding interactions and molecular assemblies of the PVR family receptors and ligands have been assessed in vitro. Furthermore, the anti-TIGIT monoclonal antibody BMS-986207 crystal structure in complex with TIGIT was determined and shows that the antibody binds an epitope that is commonly targeted by the CD155 ligand as well as other clinical anti-TIGIT antibodies. In contrast to previously proposed models, where TIGIT outcompetes costimulatory receptor CD226 for binding to CD155 due to much higher affinity (nanomolar range), our data rather suggest that PVR family members all engage in interactions with relatively weak affinity (micromolar range), including TIGIT and CD155 interactions. Thus, TIGIT and other PVR inhibitory receptors likely elicit immune suppression via increased surface expression rather than inherent differences in affinity. This work provides an improved foundational understanding of the PVR family network and mechanistic insight into therapeutic antibody intervention.
Introduction
Immune checkpoints are known to play an essential role in the regulation of immune homeostasis. Since the discovery of co-inhibitory receptors CTLA-4 and PD-1 and the success of checkpoint blockade in clinics, there has been tremendous interest in the development of next-generation checkpoint inhibitors against targets such as Lag-3, Tim3, TIGIT, CD96, NKG2A, and more.Citation1–3 TIGIT is a transmembrane glycoprotein receptor consisting of an extracellular immunoglobulin variable domain (IgV) and intracellular immunoreceptor tyrosine tail (ITT)-like and immunoreceptor tyrosine-based inhibitory (ITIM) motifs in its cytoplasmic domain.Citation4 TIGIT participates in a complex poliovirus receptor (PVR)-like protein signaling network consisting of CD96 (TACTILE), CD112R (PVRIG), CD226 (DNAM-1), CD155 (PVR/NECL-5), and CD112 (Nectin-2/PVRL2) that plays a critical role in regulating innate and adaptive immunity.Citation4,Citation5 TIGIT expression has been observed on activated CD8+ and CD4+ T cells, regulatory T (T-reg) cells, follicular T-helper (T-fh) cells, and natural killer (NK) cells in humans and is found upregulated in tumor-infiltrating lymphocytes (TILs) associated with various cancers.Citation2–5 One of the primary ligands to TIGIT, CD155, has also been observed to be highly expressed in tumor cells and is upregulated in many solid tumor types.Citation6–8 The interaction between TIGIT and CD155 triggers inhibitory signaling and results in immune suppression, thereby contributing to immune evasion. Additionally, the interaction of CD226 and CD155 leads to stimulation and positive regulation of NK and T cells.Citation2–5
The contribution of TIGIT to immune evasion has led to its emergence as a novel target for cancer immunotherapy, and many anti-TIGIT antibodies are being developed and evaluated in the clinic at this time. In particular, BMS-986207 is a human monoclonal antibody with inert fragment crystallizable (Fc) that is currently evaluated as monotherapy and in combination for advanced solid tumors. BMS-986207 is designed to disrupt the interaction of TIGIT with its ligands, block downstream signaling, and make CD155 available to interact with the costimulatory receptor CD226.
In this study, we characterized monovalent affinities and binding stoichiometry of extracellular domains of PVR family members by surface plasmon resonance (SPR) and multi-angle light scattering coupled with size-exclusion chromatography (SEC-MALS). Furthermore, the crystal structure of BMS-986207 in complex with TIGIT was determined. To our knowledge, the structure reported here (Protein Data Bank accession code 8SZY) is the first reported structure of a clinical anti-TIGIT antibody bound onto TIGIT. These results provide novel insight into PVR family receptor–ligand interactions and elucidate the molecular basis of TIGIT blockade by the BMS-986207 monoclonal antibody.
Results
Monovalent affinities and binding stoichiometries of PVR family receptors and ligands
To investigate the interactions between the PVR family members, the binding of the four receptors (TIGIT, CD112R, CD96, and CD226) to the two ligands (CD155 and CD112) were assessed by SPR (). Full-length extracellular domains of the ligands and receptors, with the exception of the CD96 (D1 only), were produced with a C-terminal hexahistidine as a tag for purification. An additional C-terminal AviTag was fused to the receptors for capture on biosensors. TIGIT and CD226 were able to bind to both CD155 and CD112 ligands. However, CD96 and CD112R were only bound to CD155 and CD112, respectively. In general, these monovalent receptor–ligand interactions are relatively weak, ranging from triple-digit nanomolar to single-digit micromolar steady-state binding constants. These determinations are consistent with some previously reported monovalent affinities.Citation9 However, they differ from some other previously reported values, which were measured using different methods and may have been influenced by avidity.Citation4,Citation10
Table 1. Monovalent SPR binding kinetics of PVR family members at steady-state equilibrium.
To gain insight into the molecular assemblies of the individual receptors and ligands and their complexes, their binding stoichiometries were assessed by SEC-MALS (). All receptors and ligands were monomeric in solution with the exception of CD96, which was observed to be a homodimer, as recently described.Citation11 By SEC-MALS, TIGIT and CD112 were observed to be monomers, although they were previously shown to be homodimers in solution by analytical ultracentrifugation and gel filtration methods.Citation12,Citation13 The molar masses of the complexes all indicate the formation of a 1:1 heterodimeric complex. While these results agree with the molecular assemblies of the CD96-CD155 and CD226-CD155 complex crystal structures, they differ from the 2:2 binding mode between TIGIT and the CD155 and CD112 ligands,Citation9,Citation14–16 where the membrane distal domains of the receptors and ligands join together and reveal a common recognition organization of the PVR family members.
Table 2. SEC-MALS of PVR family ligand and receptor complexes as well as Fab complexes with TIGIT. The individual Fab molecular weights are approximately 48 kDa. All proteins have full-length extracellular Ig domains, with the exception of CD96 (D1 only) being truncated.
To further investigate the differences in binding assemblies, the structures of the ligands and receptors were inspected. TIGIT has been reported to form homodimers, but its homodimerization interface has been identified to be on the opposite face of the ligand-binding site as compared to CD96.Citation13,Citation17 While the ligand-binding interface of CD96 is occluded as a homodimer and must dissociate to allow each protomer to engage ligand in a 1:1 binding stoichiometry (),Citation11 TIGIT can readily form complexes with CD155 since its ligand-binding site is accessible (). Indeed, TIGIT-CD155 interactions have a faster association rate than CD96-CD155 (Figure S1). These molecular assembly differences between TIGIT and CD96 may have underlying mechanistic and biological consequences.
Figure 1. Schematic of binding assemblies of CD155 with monomeric and homodimeric CD96 and TIGIT.
Crystal structure of BMS-986207 in complex with TIGIT
To investigate the molecular basis of recognition of BMS-986207 to TIGIT, the complex crystal structure was determined. Many different molecular formats were screened for crystallization, such as by using anti-Fab crystallization chaperones or replacing the human constant domains with that of a mouse. However, these attempts all yielded poorly diffracting crystals (~8 Å), which were not amenable to structure determination. Ultimately, a non-blocking antibody, CHA.9.543,Citation18 was used to create a ternary complex with BMS-986207 and TIGIT, which formed crystals that diffracted to a nominal resolution of ~2.7 Å (, ). Two independent copies of the complex in the asymmetric unit present a similar overall organization and superpose with a Cα-root-mean-square deviation (Cα-RMSD) of 0.55 Å. Superposing the two copies of TIGIT yielded a Cα-RMSD of 0.23 Å. The calculated Cα-RMSD of BMS-986207 Fab and CHA.9.543 Fab were 0.32 Å and 0.33 Å. Superposition of TIGIT molecule extracted from the solution and previously published x-ray crystal structure (PDB ID 3UDW and PDB ID 5V52) had a Cα-RMSD of 0.34 Å and 0.40 Å, respectively, suggesting no significant structural differences occurred upon the binding of anti-TIGIT antibodies, CD155, and CD112 (Figure S2).
Figure 2. Crystal structure of BMS-986207 and CHA.9.543 in complex with TIGIT. (a) The crystal structure of BMS-986207:TIGIT:CHA.9.543 reveals a heterotrimeric assembly with a single TIGIT sandwiched by the two anti-TIGIT antibodies. TIGIT is colored in light pink and represented by ribbon and surface. BMS-986207 is represented in blue and cyan ribbons, whereas CHA.9.543 is represented in gray ribbons. All antibody light chains were represented by a lighter hue and heavy chains were represented by a darker shade. (b) Detailed view of the light and heavy chain complementarity-determining region (CDR) at the TIGIT binding interface. Interacting residues from each of the CDRs are displayed as sticks. (c) Epitope of TIGIT engaged by BMS-986207 are shown in surface representation, with heavy chain contribution colored in blue and light chain contribution colored in cyan.
Table 3. X-ray diffraction data collection and refinement statistics.
The crystal structure unveils that BMS-986207 shares an overlapping epitope on TIGIT with CD155, confirming that BMS-986207 is a direct blocker of ligand interactions (). Conversely, CHA.9.543 binds on the opposing face of TIGIT from where BMS-986207 and CD155 recognize. BMS-986207 Fab utilizes four complementarity-determining regions (CDRs) (HCDR1, HCDR3, LCDR1, and LCDR2) to contact TIGIT, with heavy chain and light chain contributing 71% and 29% of the Fab buried surface area, respectively (). While CD155 buries ~800 Å2 on TIGIT, BMS-986207 binds a larger footprint (~950 Å2). Interestingly, BMS-986207 has a long 20-residue HCDR3, accounting for the majority of the paratope (65%), and contributes to the majority of the epitope on TIGIT. On the other hand, CHA.9.543 buries ~908 Å2 on TIGIT and its heavy and light chain contribute to 57% and 43% of the Fab buried surface area, respectively (Figure S3).
Figure 3. TIGIT recognition comparisons by BMS-986207 and natural ligands. The TIGIT AX6G lock, shown as sticks, is occupied similarly by (a) CD155 F128 (PDB ID 3UDW), (b) BMS-986207 HCDR3 F100j, (c) CD112 F145 (PDB ID 5V52), and (d) MG1131 HCDR3 W100 (PDB ID 7VYT).
Common TIGIT recognition motifs by antibodies and ligands
PVR family homo- or hetero-dimerization can be defined by a pair of “lock-and-key” binding motifs, where a hydrophobic pocket formed by an AX6G peptide (lock) of one protein interacts with an aromatic residue from a T(Y/F)P tripeptide motif (key) of the other protein.Citation9,Citation10,Citation14,Citation17 For instance, these recognition motifs have been observed in the crystal structures of TIGIT bound to its ligands.Citation9,Citation17 Interestingly, the F100j residue in the HCDR3 of BMS-986207 (Kabat numbering) extends into the TIGIT lock and mimics that of F128 of CD155 or F145 of CD112 on the TIGIT ligand complexes (). To evaluate the importance of this residue for TIGIT binding, the F100j amino acid was substituted to alanine, arginine, or tyrosine. Complete loss of binding was observed with the F100j-R variant and significantly weaker binding was observed with F100j-A (). The F100j-Y substitution had similar binding compared to the parental antibody, which indicates that an aromatic residue is important at this position for BMS-986207 recognition to TIGIT.
Figure 4. Importance of BMS-986207 HCDR3 residue F100j for TIGIT recognition by site-directed mutagenesis. The binding of Fab variants (50 nM) to TIGIT was measured by BLI.
In a recent report, anti-TIGIT antibody MG1131 also showed similar molecular mimicry, where HCDR3 residue W100 (Kabat numbering) inserts into the TIGIT lock ().Citation19 Significant binding loss was also observed when the tryptophan residue was mutated to an alanine.Citation19 The utilization of key and mimicking native ligands is not unique to TIGIT antibodies and has been described also for antibody against CD96, another member of the PVR family.Citation11 These observations suggest that the lock-and-key binding motif is a common means for TIGIT recognition that is not only present in native ligands but is also utilized by multiple antibodies that target the PVR family.
Anti-TIGIT antibodies bind a similar epitope and directly block CD155 interactions
To investigate where BMS-986207 and other clinical anti-TIGIT monoclonal antibodies bind TIGIT, reported epitopes from patent literatureCitation18,Citation20–22 were compared to the footprint of CD155. Comparative models reveal that all antibodies converge on a similar epitope for TIGIT recognition to prevent simultaneous CD155 binding (). Furthermore, the SPR kinetics of the anti-TIGIT antibodies to human TIGIT interactions were measured and revealed that they have a wide range of affinities, spanning three orders of magnitude (). For instance, anti-TIGIT A had the highest affinity (0.075 nM), while anti-TIGIT D had the weakest affinity (40 nM). Using SEC-MALS, the Fab-TIGIT complexes had an observed molar mass of ~65 kDa (), suggesting that they all formed 1:1 Fab-TIGIT complex. These findings collectively show that these tested clinical anti-TIGIT monoclonal antibodies bind to the same TIGIT epitope, albeit with different affinities, and all compete with CD155 ligand.
Figure 5. Binding epitope of CD155 and antibody footprints on TIGIT. The footprints of (a) BMS-986207, (b) Tiragolumab,Citation19 (c) Vibostolimab,Citation20 and (d) DomvanalimabCitation21 are highlighted in blue, green, yellow, and hot pink. TIGIT is represented as pink ribbon and the CD155 epitope is displayed as orange surface in all models. (e) the epitope residues for each antibody are color coded in the human TIGIT ECD sequence.
Table 4. SPR binding kinetics of anti-human TIGIT antibodies.
Discussion
The extracellular domain of TIGIT has been previously reported to form a homodimer in solution. The role of cell surface TIGIT cis-homodimerization for cell adhesion and CD155 downstream signaling was also reported.Citation10 In contrast, we observed that under the tested conditions the isolated extracellular domain of TIGIT exists as a monomer by SEC-MALS analysis. It is possible that the membrane-bound TIGIT does homodimerize on cell surface at high local concentrationsCitation13,Citation17 and at high concentrations required for crystallization of the TIGIT:CD155 or TIGIT:CD112 complexes.Citation9,Citation17 In our crystal structure, the chaperone anti-TIGIT CHA.9.543 antibody binds to the predicted TIGIT dimer interface to prevent the putative TIGIT homodimerization (Figure S3).
A widely held model was put forward where inhibitory TIGIT:CD155 high-affinity interactions (nM) outcompete the lower affinity costimulatory CD155:CD226 complex (μM).Citation4 This model, however, was based on Fc-fused ligand affinity which leads to avidity and was compared to monomeric interaction of CD226:CD155. In our hands, when the receptor and ligand-binding interactions were measured in a monovalent format, they all interacted with much weaker affinity and had dissociation rate constants ranging from 270 nM to 3,700 nM. CD155 had the highest affinity to CD226 (530 nM), followed by TIGIT (1,150 nM), and lastly CD96 (3,700 nM). In contrast to previously reported affinities, CD155 bound CD226 and TIGIT with similar affinity and is suggestive of an alternative mechanism of inhibition where surface expression of comparable affinity TIGIT may be utilized to modulate CD226 activity.Citation23,Citation24 This mechanism and affinity range of CD155:TIGIT is comparable to other checkpoint inhibitors (such as PD1:PDL1, CD80:CTLA4), which also have μM affinities.Citation25,Citation26 Comparable and low affinities of inhibitory and stimulatory ligands provide more reversibility in the context of cell–cell interaction, as well as amenable modulation via expression levels and antibody intervention.
SPR kinetics study of clinical anti-TIGIT antibodies (BMS-986207, anti-TIGIT A, anti-TIGIT B, anti-TIGIT C, anti-TIGIT D, and anti-TIGIT E) revealed a range of monovalent affinities with varying association and dissociation rates. Epitope information determined using various methods collectively demonstrate that all tested clinical anti-TIGIT monoclonal antibodies bind a very similar epitope. It is important to note that clinical anti-TIGIT antibodies have a variety of backbones (e.g., hIgG1 wild-type, Fc-inert, and ADCC-enhanced) that may affect the potency and clinical efficacies.Citation27–29 It will be interesting to monitor the clinical progress of anti-TIGIT antibodies and see if differences in activities can be correlated with affinity or Fc effector function differences.
We observed a conserved “lock-and-key” binding motif at the binding interface of BMS-986207:TIGIT that mimics a previously observed motif at the CD155:TIGIT interface. Intriguingly, another anti-TIGIT antibody, MG1131, similarly used lock-and-key binding motifs for TIGIT recognition.Citation24 Furthermore, we can infer the AX6G lock motif appears to be a part of clinical anti-TIGIT A and B’s footprints on TIGIT. These observations may suggest that the hydrophobic lock and aromatic residue key binding motifs are common solutions for therapeutic antibodies aiming to bind TIGIT and simultaneously block CD155.
Materials and methods
Protein expression and purification
The Fab constructs were cloned into the National Research Council of Canada (NRC)’s proprietary pTT5 backbone by GenScript. The Fab constructs have an N-terminal osteonectin signal peptide and a hexahistidine tag fused to the C-terminus of the heavy chain. Human TIGIT, CD226, CD96, CD155, CD112, and CD112R were similarly cloned into the pTT5 backbone with a C-terminal hexahistidine tag and BirA biotinylation tag. For crystallography studies, a human TIGIT construct with C47S mutation and only a C-terminal hexahistidine tag was also generated. The proteins were expressed in Expi293F cells (Invitrogen). Supernatants were harvested on post-transfection day 5–7 and subjected to Ni Sepharose excel (Cytiva), and buffer exchanged into 1× PBS for −80°C storage.
X-ray crystallography and structural analyses
Human TIGIT was expressed using Expi293F cells (Invitrogen) treated with kifunensine. Purified protein was treated with Endoglycosidase H. For complex formation, BMS-986207 and CHA.9.543 Fabs were added to TIGIT in molar excess and incubated at room temperature for 1 hour. The ternary complex was purified from excess Fabs by gel filtration in 50 mM NaCl, 10 mM Tris pH 8.0 buffer. Crystals of the complex were grown by sitting drop vapor diffusion at 20°C by mixing 0.2 µL of concentrated protein sample (17.5 mg/mL) with 0.2 µL of mother liquor (0.2 M sodium phosphate monobasic monohydrate, 20% w/v Polyethylene glycol 1,000). Crystals were cryo-protected with 15% ethylene glycol and flash-cooled in liquid nitrogen. X-ray diffraction data were collected at beamline 17-1D (wavelength 1.0 Å) at the Advanced Photon Source (Argonne National Laboratory, Lemont, IL) under cryo conditions using a DECTRIS Eiger2 X 9 M detector. Crystals diffracted to a nominal resolution of 2.725 Å resolution, with anisotropic diffraction limits of 2.923 Å, 2.326 Å, and 2.306 Å. Diffraction data were processed with the autoPROCCitation30 toolbox that made use of external programs XDS/XSCALE,Citation31 POINTLESS,Citation32 CCP4,Citation33 and STARANISO (Global Phasing Limited) for ellipsoidal truncation and anisotropic scaling. The complex was determined by molecular replacement with Phaser using PDB ID 4NM4 for the Fab constant domains of both BMS-986207 and CHA.9.543, PDB ID 6B3S (HCDR3 removed) for the BMS-986207 variable domain, PDB ID 1PLG for the variable domain of CHA.9.543, and 3UDW for TIGIT. Two ternary complexes were found in the asymmetric unit. The model was iteratively built using CootCitation15 and refined in Phenix.Citation16 In the final structure, 96.64% of the residues are in favored regions of the Ramachandran plot with 0.21% outliers, as calculated by MolProbity.Citation34 X-ray diffraction data collection and refinement statistics are reported in . Areaimol and PISA of the CCP4Citation33 suite was used to calculate the molecular epitope and paratope. PyMOL was used to render structure models and figures. Kabat numbering was applied to the Fabs. Coordinates and structure factors have been deposited to the RCSB Protein Data Bank under the accession code 8SZY.
Biolayer interferometry binding measurements
Binding interactions between Fabs and TIGIT were assessed by biolayer interferometry using an Octet® HTX instrument. Before the start of the experiment, streptavidin biosensors were hydrated for 10 minutes in a baseline buffer (1× phosphate-buffered saline (PBS), 0.05% Tween 20, and 0.5% bovine serum albumin (BSA)). Then, 10 μg/mL BirA biotinylated human TIGIT was loaded for three minutes, and then a baseline was established for 30 seconds in buffer only. Subsequently, the sensors were placed into a well containing Fabs at 50 nM and 500 nM for 3 minutes (association), followed by buffer only for 5 minutes (dissociation).
Size exclusion chromatography with multi-angle light scattering
TIGIT, CD155, and anti-TIGIT Fab complexes were formed with 2.5-fold molar excess TIGIT antigens and normalized to 2 mg/mL. Free receptors and ligands were normalized to 1 mg/mL. Forty micrograms of protein samples was injected onto Acquity UPLC Protein BEH SEC columns (200 Å, 1.7 μm, 4.6 mm × 300 mm and 125 Å, 1.7 μm, 4.6 mm × 300 mm, Waters) attached to an Acquity UPLC H-Class system (Waters) at an isocratic flow rate of 0.3 mL/min in 1× PBS, 0.05% sodium azide. The eluted peaks were analyzed using a UV detector operated at 280 nm wavelength followed by the DAWN HELEOS-II/µDAWN multiangle light scattering detector and an Optilab T-rEX differential refractometer (Wyatt Technology). The molar mass distribution of the proteins was calculated in the Astra v7.3 software using the Zimm model, with a dn/dc refractive index increment value of 0.185. For all proteins and complexes, protein conjugate analysis was conducted using MALS-UV-RI. A dn/dc value of 0.138 mL/g was used for glycans as the modifier.
SPR determination of binding parameters via BIACORE
SPR was used to determine binding constants for the Fabs to human TIGIT with a BIACORE® T200 SPR spectrometer (Biacore AB, Uppsala, Sweden). Human TIGIT was captured on a CM4 anti-Avi chip and was then bound by a concentration series of the Fabs (500 nM, 100 nM, 20 nM, 4 nM, and 0.8 nM) in HBS pH 7.4 running buffer supplemented with 0.05% Tween-20, 1 g/L BSA at 37°C. All data were double-referenced and fitted to a 1:1 Langmuir binding model with mass transport to determine equilibrium dissociation constants (KD) as well as association (ka) and dissociation (kd) rate constants.
SPR determination of binding parameters via ProteOn
A ProteOn XPR36 (Bio-Rad) surface plasmon resonance instrument was used to measure the affinities of PVR receptors to their ligands at 37°C. Each of the four biotinylated receptors (human TIGIT, CD112R, CD226, and CD96 D1) was diluted to ~4 µg/mL in running buffer (HBST with 0.01% BSA) and immobilized via NeutrAvidin capture to unique “vertical” flow channels of a ProteOn NLC biosensor chip (Bio-Rad) at levels ranging from approximately 400 RU to 425 RU. In the horizontal flow direction, each ligand (human CD155, and CD112) was flowed as an analyte over all four receptors using unique injection ports simultaneously in a two-fold dilution series at a concentration range of ~10 nM to 20 µM. Association times were 2 minutes followed by 5 minutes of dissociation at a flow rate of 50 µL/min. NeutrAvidin interspots were used as reference surfaces on the NLC chip while several blank injections were averaged for double referencing. Sensorgram data was processed using Scrubber P43 software (BioLogic Software). No regeneration of the surfaces was needed because of short dissociation times. Both the association and dissociation of the sensorgrams were too rapid for reliable kinetic measurements, therefore the KD values were estimated using a steady-state equilibrium fitting model using Scrubber P43 software.
Abbreviations
| ADCC | = | antibody-dependent cellular cytotoxicity |
| Cα-RMSD | = | Cα-root-mean-square deviation |
| CDRs | = | complementarity-determining regions |
| Fab | = | fragment antigen binding |
| Fc | = | fragment crystallizable |
| IgG | = | immunoglobulin G |
| ITIM | = | immunoreceptor tyrosine-based inhibitory |
| ITT | = | immunoreceptor tyrosine tail |
| NK | = | natural killer cell |
| NRC | = | National Research Council of Canada |
| PVR | = | poliovirus receptor |
| RU | = | response unit |
| SEC-MALS | = | size exclusion chromatography with multi-angle light scattering |
| SPR | = | surface plasmon resonance |
| T-fh | = | follicular T helper cell |
| TIL | = | tumor-infiltrating lymphocyte |
| T-reg | = | regulatory T cell |
Supplemental Material
Download MS Word (1.2 MB)Supplemental Material
Download MS Word (3.4 MB)Acknowledgments
The authors thank Olin Chang, Amanda Rhea, Megan Smith, Julie Zorn, and Bryant Chau for reagents, technical advice, and scientific discussions. The authors thank Steven Sheriff, Jodi Muckelbauer, and the staff at the IMCA-CAT at the Argonne National Laboratory for assistance in x-ray data collection and beamline support.
Disclosure statement
All authors are current or former employees of Bristol Myers Squibb.
Supplementary material
Supplemental data for this article can be accessed online at https://doi.org/10.1080/19420862.2023.2253788
Additional information
Funding
References
- Korman AJ, Peggs KS, Allison JP. Checkpoint blockade in cancer immunotherapy. Adv Immunol. 2006;90:297–10. doi:10.1016/S0065-2776(06)90008-X. PMID: 16730267.
- Ge Z, Peppelenbosch MP, Sprengers D, Kwekkeboom J. TIGIT, the next step towards successful combination immune checkpoint therapy in cancer. Front Immunol. 2021;12:699895. doi:10.3389/fimmu.2021.699895. PMID: 34367161.
- Anderson AC, Joller N, Kuchroo VK. Lag-3, Tim-3, and TIGIT: Co-inhibitory receptors with specialized functions in immune regulation. Immunity. 2016;44(5):989–1004. doi:10.1016/j.immuni.2016.05.001. PMID: 27192565.
- Yu X, Harden K, Gonzalez LC, Francesco M, Chiang E, Irving B, Tom I, Ivelja S, Refino CJ, Clark H, et al. The surface protein TIGIT suppresses T cell activation by promoting the generation of mature immunoregulatory dendritic cells. Nat Immunol. 2009;10:48–57. doi:10.1038/ni.1674. PMID: 19011627.
- Boles KS, Vermi W, Facchetti F, Fuchs A, Wilson TJ, Diacovo TG, Cella M, Colonna M. A novel molecular interaction for the adhesion of follicular CD4 T cells to follicular DC. Eur J Immunol. 2009;39(3):695–703. doi:10.1002/eji.200839116. PMID: 19197944.
- Casado JG, Pawelec G, Morgado S, Sanchez-Correa B, Delgado E, Gayoso I, Duran E, Solana R, Tarazona R. Expression of adhesion molecules and ligands for activating and costimulatory receptors involved in cell-mediated cytotoxicity in a large panel of human melanoma cell lines. Cancer Immunol Immunother. 2009;58(9):1517–26. doi:10.1007/s00262-009-0682-y. PMID: 19259667.
- Masson D, Jarry A, Baury B, Blanchardie P, Laboisse C, Lustenberger P, Denis MG. Overexpression of the CD155 gene in human colorectal carcinoma. Gut. 2001;49(2):236–40. doi:10.1136/gut.49.2.236. PMID: 11454801.
- Nakai R, Maniwa Y, Tanaka Y, Nishio W, Yoshimura M, Okita Y, Ohbayashi C, Satoh N, Ogita H, Takai Y, et al. Overexpression of Necl-5 correlates with unfavorable prognosis in patients with lung adenocarcinoma. Cancer Sci. 2010;101(5):1326–30. doi:10.1111/j.1349-7006.2010.01530.x. PMID: 20331633.
- Deuss FA, Gully BS, Rossjohn J, Berry R. Recognition of nectin-2 by the natural killer cell receptor T cell immunoglobulin and ITIM domain (TIGIT). J Biol Chem. 2017;292:11413–22. doi:10.1074/jbc.M117.786483. PMID: 28515320.
- Saha S, Sparkes A, Matus EI, Lee P, Gariepy J. The IgV domain of the poliovirus receptor alone is immunosuppressive and binds to its receptors with comparable affinity. Sci Rep. 2023;13(1):4609. doi:10.1038/s41598-023-30999-w. PMID: 36944702.
- Lee PS, Chau B, Barman I, Bee C, Jashnani A, Hogan JM, Aguilar B, Dollinger G, Rajpal A, Strop P. Antibody blockade of CD96 by distinct molecular mechanisms. MAbs. 2021;13(1):1979800. doi:10.1080/19420862.2021.1979800. PMID: 34595996.
- Harrison OJ, Vendome J, Brasch J, Jin X, Hong S, Katsamba PS, Ahlsen G, Troyanovsky RB, Troyanovsky SM, Honig B, et al. Nectin ectodomain structures reveal a canonical adhesive interface. Nature Structural & Molecular Biology. 2012;19(9):906–15. doi:10.1038/nsmb.2366. PMID: 22902367.
- Samanta D, Guo H, Rubinstein R, Ramagopal UA, Almo SC. Structural, mutational and biophysical studies reveal a canonical mode of molecular recognition between immune receptor TIGIT and nectin-2. Mol Immunol. 2017;81:151–59. doi:10.1016/j.molimm.2016.12.003. PMID: 27978489.
- Deuss FA, Watson GM, Fu Z, Rossjohn J, Berry R. Structural basis for CD96 immune receptor recognition of nectin-like Protein-5, CD155. Structure. 2019;27(2):219–28.e3. doi:10.1016/j.str.2018.10.023. PMID: 30528596.
- Liu J, Qian X, Chen Z, Xu X, Gao F, Zhang S, Zhang R, Qi J, Gao GF, Yan J. Crystal structure of cell adhesion molecule nectin-2/CD112 and its binding to immune receptor DNAM-1/CD226. J Immunol. 2012;188(11):5511–20. doi:10.4049/jimmunol.1200324. PMID: 22547693.
- Wang H, Qi J, Zhang S, Li Y, Tan S, Gao GF. Binding mode of the side-by-side two-IgV molecule CD226/DNAM-1 to its ligand CD155/Necl-5. Proc Natl Acad Sci U S A. 2019;116:988–96. doi:10.1073/pnas.1815716116. PMID: 30591568.
- Stengel KF, Harden-Bowles K, Yu X, Rouge L, Yin J, Comps-Agrar L, Wiesmann C, Bazan JF, Eaton DL, Grogan JL. Structure of TIGIT immunoreceptor bound to poliovirus receptor reveals a cell–cell adhesion and signaling mechanism that requires cis-trans receptor clustering. Proc Natl Acad Sci U S A. 2012;109(14):5399–404. doi:10.1073/pnas.1120606109. PMID: 22421438.
- White M, Kumar S, Chan C, Liang S, Stapleton L, Drake A, Gozlan Y, Vaknin I, Sameah-Greenwald S, Dassa L, et al. Inventor; anti-TIGIT Antibodies, Anti-PVRIG antibodies and combinations there of. 2018. WO 2018/033798 A1.
- Jeong BS, Nam H, Lee J, Park HY, Cho KJ, Sheen JH, Song E, Oh M, Lee S, Choi H, et al. Structural and functional characterization of a monoclonal antibody blocking TIGIT. MAbs. 2022;14(1):2013750. doi:10.1080/19420862.2021.2013750. PMID: 35090381.
- Grogan JL, Johnston RJ, Wu Y, Liang W, Lupardus P, Yadav M, Seshasayee D, Hazen M. Inventor; Anti-TIGIT antibodies and methods of use. 2017. WO2017053748A2.
- Williams S, Laface D, Fayadat-Dilman L, Raghunathan R, Liang L, Seghezzi W. Inventor; anti-TIGIT antibodies. 2016. WO/2016/028656.
- Tso Y, Tsurushita N, Duramad O. Inventor; ANTIBODIES to TIGIT. 2017. WO/2017/152088A1.
- Chauvin JM, Pagliano O, Fourcade J, Sun Z, Wang H, Sander C, Kirkwood JM, Chen TH, Maurer M, Korman AJ, et al. TIGIT and PD-1 impair tumor antigen–specific CD8+ T cells in melanoma patients. J Clin Invest. 2015;125(5):2046–58. doi:10.1172/JCI80445. PMID: 25866972.
- Jin HS, Park Y. Hitting the complexity of the TIGIT-CD96-CD112R-CD226 axis for next-generation cancer immunotherapy. BMB Rep. 2021;54(1):2–11. PMID: 33298247. https://www.ncbi.nlm.nih.gov/pubmed/33298247.
- van der Merwe PA, Bodian DL, Daenke S, Linsley P, Davis SJ. CD80 (B7-1) binds both CD28 and CTLA-4 with a low affinity and very fast kinetics. J Exp Med. 1997;185(3):393–404. doi:10.1084/jem.185.3.393.
- Collins AV, Brodie DW, Gilbert RJ, Iaboni A, Manso-Sancho R, Walse B, Stuart DI, van der Merwe PA, Davis SJ. The interaction properties of costimulatory molecules revisited. Immunity. 2002;17(2):201–10. doi:10.1016/s1074-7613(02)00362-x. PMID: 12196291.
- Chen X, Song X, Li K, Zhang T. FcgammaR-Binding is an important functional attribute for immune checkpoint antibodies in cancer immunotherapy. Front Immunol. 2019;10:292. doi:10.3389/fimmu.2019.00292. PMID: 30863404.
- Wang X, Mathieu M, Brezski RJ. IgG Fc engineering to modulate antibody effector functions. Protein Cell. 2018;9(1):63–73. doi:10.1007/s13238-017-0473-8. PMID: 28986820.
- Yeo J, Ko M, Lee DH, Park Y, Jin HS. TIGIT/CD226 axis regulates anti-tumor Immunity. Pharmaceuticals (Basel). 2021;14(3):200. doi:10.3390/ph14030200. PMID: 33670993.
- Vonrhein C, Flensburg C, Keller P, Sharff A, Smart O, Paciorek W, Womack T, Bricogne G. Data processing and analysis with the autoPROC toolbox. Acta Crystallogr D Biol Crystallogr. 2011;67(4):293–302. doi:10.1107/S0907444911007773. PMID: 21460447.
- Kabsch W. Xds. Acta Crystallogr D Biol Crystallogr. 2010;66(2):125–32. doi:10.1107/S0907444909047337. PMID: 20124692.
- Evans P. Scaling and assessment of data quality. Acta Crystallogr D Biol Crystallogr. 2006;62(1):72–82. doi:10.1107/S0907444905036693. PMID: 16369096.
- Winn MD, Ballard CC, Cowtan KD, Dodson EJ, Emsley P, Evans PR, Keegan RM, Krissinel EB, Leslie AG, McCoy A, et al. Overview of the CCP4 suite and current developments. Acta Crystallogr D Biol Crystallogr. 2011;67(4):235–42. doi:10.1107/S0907444910045749. PMID: 21460441.
- Chen VB, Arendall WB 3rd, Headd JJ, Keedy DA, Immormino RM, Kapral GJ, Murray LW, Richardson JS, Richardson DC. MolProbity: all-atom structure validation for macromolecular crystallography. Acta Crystallogr D Biol Crystallogr. 2010;66(1):12–21. doi:10.1107/S0907444909042073. PMID: 20057044.