Structural basis for cancer immunotherapy by the first-in-class checkpoint inhibitor ipilimumab

Udupi A. Ramagopal, Weifeng Liu, Sarah C. Garrett-Thomson, Jeffrey B. Bonanno, Qingrong Yan, Mohan Srinivasan, Susan C. Wong, Alasdair Bell, Shilpa Mankikar, Vangipuram S. Rangan, Shrikant Deshpande, Alan J. Korman, and Steven C. Almo
PNAS May 23, 2017 114 (21) E4223-E4232; first published May 8, 2017 https://doi.org/10.1073/pnas.1617941114

  1. Edited by James P. Allison, MD Anderson Cancer Center, University of Texas, Houston, TX, and approved April 6, 2017 (received for review November 1, 2016)

Significance

Biologics represent a major class of therapeutics for the treatment of malignancies, autoimmune diseases, and infectious diseases. Ipilimumab is the first-in-class immunotherapeutic for blockade of CTLA-4 and significantly benefits overall survival of patients with metastatic melanoma. The X-ray crystal structure of the ipilimumab:CTLA-4 complex defines the atomic interactions responsible for affinity and selectivity and demonstrates that the therapeutic action of ipilimumab is due to direct steric competition with the B7 ligands for binding to CTLA-4.

Abstract

Rational modulation of the immune response with biologics represents one of the most promising and active areas for the realization of new therapeutic strategies. In particular, the use of function blocking monoclonal antibodies targeting checkpoint inhibitors such as CTLA-4 and PD-1 have proven to be highly effective for the systemic activation of the human immune system to treat a wide range of cancers. Ipilimumab is a fully human antibody targeting CTLA-4 that received FDA approval for the treatment of metastatic melanoma in 2011. Ipilimumab is the first-in-class immunotherapeutic for blockade of CTLA-4 and significantly benefits overall survival of patients with metastatic melanoma. Understanding the chemical and physical determinants recognized by these mAbs provides direct insight into the mechanisms of pathway blockade, the organization of the antigen–antibody complexes at the cell surface, and opportunities to further engineer affinity and selectivity. Here, we report the 3.0 Å resolution X-ray crystal structure of the complex formed by ipilimumab with its human CTLA-4 target. This structure reveals that ipilimumab contacts the front β-sheet of CTLA-4 and intersects with the CTLA-4:Β7 recognition surface, indicating that direct steric overlap between ipilimumab and the B7 ligands is a major mechanistic contributor to ipilimumab function. The crystallographically observed binding interface was confirmed by a comprehensive cell-based binding assay against a library of CTLA-4 mutants and by direct biochemical approaches. This structure also highlights determinants responsible for the selectivity exhibited by ipilimumab toward CTLA-4 relative to the homologous and functionally related CD28.

Activation of the immune system to target and eliminate malignancies is recognized as one of the most promising directions for cancer therapy (1⇓⇓–4). Two broad strategies for immunotherapy may be envisaged: inhibition of negative regulators of immune responsiveness (collectively known as checkpoint blockade) (2, 5⇓⇓–8) and activation of costimulatory pathways (8). A powerful example is provided by antibodies targeting cytotoxic T lymphocyte-associated antigen 4 (CTLA-4), a T-cell surface molecule, which like the homologous CD28 (∼30% sequence identity) binds the B7-1 and B7-2 ligands (9). While CD28 is constitutively expressed and is required, in conjunction with TCR engagement, for T-cell activation, CTLA-4 is a negative regulator of T-cell function expressed after T-cell activation to terminate the response. Ipilimumab, a fully human antibody targeting CTLA-4 (marketed as Yervoy), demonstrated improved overall survival in two phase-III clinical trials of metastatic melanoma (10, 11) and received FDA approval for the treatment of metastatic melanoma in 2011. Ipilimumab is the first-in-class immunotherapeutic for blockade of CTLA-4 and significantly benefits overall survival of patients with metastatic melanoma. Notably, combination therapies involving ipilimumab and other immunomodulators/checkpoint antibodies, such as those targeting PD-1 or tremelimumab (anti–CTLA-4) and anti–PD-L1, can result in enhanced activity (12⇓⇓–15).

Multiple mechanisms have been described for CTLA-4 function. These include negative signals from intrinsic effects of CTLA-4 expressed on activated T effector cells as a result of ligand binding. Extrinsic effects of CTLA-4 are the consequence of the ability of CTLA-4 expressed on Tregs to remove B7 ligands from the surface of dendritic cells or antigen presenting cells, resulting in significantly reduced suppression (16). This regulatory mechanism of transendocytosis may also be operative in activated T effector cells (17, 18). CTLA-4–B7 interactions also have intrinsic effects on Treg, dampening their proliferation or activation (19).

Antibodies to CTLA-4 also operate through multiple mechanisms. Recent studies in murine models suggest that antibodies targeting CTLA-4 delete intratumoral Treg cells through an Fcγ receptor (FcγR)-dependent process (20, 21). Although Treg depletion does not require ligand blocking, ample evidence indicates that inhibition of ligand binding to CTLA-4 is an important factor contributing to the antitumor activity of anti–CTLA-4 antibodies in murine models as well as in humans. These data include the demonstration in murine models that targeting of the effector T-cell compartment contributes to the antitumor activity of anti–CTLA-4, whereas exclusive targeting of the Treg cell compartment failed to elicit tumor protection—thus highlighting the importance of both modalities for antitumor activity (6). Blockade of CTLA-4 also promotes Treg suppression in vitro (6, 19).

Human clinical studies with tremelimumab, which, like ipilimumab, blocks the interactions of CTLA-4 with its ligands, demonstrate that this antibody also has antitumor activity in addition to inducing adverse events (22). Moreover, long-term survival of melanoma patients treated with either ipilimumab (23) or tremelimumab have been reported (24). Notably, tremelimumab harbors the IgG2 isotype and thus cannot engage FcγRs, supporting a mechanism of action that relies largely on competitive inhibition with the B7 ligands (ref. 20 and references therein). In contrast, ipilimumab is an IgG1 that can engage human FcγR; consistent with this behavior, ipilimumab was shown to mediate antibody-dependent cell-mediated cytotoxicity (ADCC)-facilitated depletion of Treg cells in vitro (25). However, only small numbers of patients have been analyzed for depletion of Treg at the tumor site by ipilimumab (26, 27). Of note, FcγR polymorphisms have no impact on the survival of ipilimumab-treated patients (28). Importantly, all aspects of ipilimumab-mediated CTLA-4 blockade require specific and high-affinity recognition of CTLA-4.

CTLA-4 and CD28 are type-I integral membrane proteins composed of a single Ig variable domain (IgV), a transmembrane segment, and a cytoplasmic tail bearing various signaling motifs. The IgV ectodomains share ∼30% sequence identity and exhibit a two-layered beta sheet involving the A′GFCC′C′′ strands of the front sheet and the ABED strands of the back sheet. Both molecules exist as covalent homodimers due to a disulfide bond formed between cysteines in the stalk segments connecting the IgV and transmembrane domains. A wide range of structural and biochemical data demonstrate that, despite this modest sequence similarity, both CTLA-4 and CD28 use the stereochemical features of a shared proline-rich motif (MYPPPY), present in the loops joining the F and G β strands, to bind the B7-1 and B7-2 ligands (9). An essential property of any CTLA-4 therapeutic antibody is the ability to specifically engage CTLA-4, while exhibiting little or no cross-reactivity with CD28. In the case of cancer immunotherapy, recognition of CD28 and inhibition of ligand binding could inhibit T-cell activation, which would oppose the desired therapeutic activity. To define the nature of the inhibitory mechanism and the specificity exhibited by ipilimumab for CTLA-4, we report the crystal structure of the complex formed by a Fab fragment of ipilimumab and human CTLA-4, as well as complementary biochemical studies that confirm the epitope recognized by ipilimumab. This work unambiguously defines the ipilimumab recognition surface on CTLA-4, which partly overlaps the B7 ligand binding surfaces, indicating that direct steric competition contributes to the function of ipilimumab. This work also highlights the determinants responsible for the highly selective binding to CTLA-4 and provides the foundation for structure-guided engineering of ipilimumab variants with new in vitro activities (e.g., altered affinities for CTLA-4 and Fc receptors) for the realization of enhanced in vivo therapeutic functions.

Results and Discussion

Overall Structure of the Human CTLA-4:Ipilimumab Complex.

The structure of the complex between monomeric human CTLA-4 (residues 1–118) and the Fab fragment derived from ipilimumab was determined and refined to a resolution of 3.0 Å, with Rwork and Rfree of 20.3% and 26.8%, respectively (Fig. 1A, Fig. S1A, and Table 1). The two independent copies of the complex in the asymmetric unit exhibit similar overall organization and superpose with a Cα–root-mean-square deviation (Cα–rmsd) of 0.74 Å, calculated over all experimentally defined Cα atoms of CTLA-4 (2–116) and the VH (2–118) and VL (2–108) domains of the ipilimumab Fab fragment (Fig. S1; the Cα–rmsd is 1.24 Å when the constant domains of the Fab are included). The Cα–rmsd between the two molecules of CTLA-4 in the asymmetric unit is 0.65 Å. The ipilimumab Fab exhibited typical CDR structural parameters, with elbow angles for the two independent molecules in the asymmetric unit of 168° and 170°. Due to the similarity of the two complexes, the discussion below refers to one of the CTLA-4:Fab complexes in the asymmetric unit (Fig. 1 and Fig. S1B), denoted L for light chain, H for heavy chain, and C for CTLA-4.

Fig. 1.
Fig. 1.

Structure of the CTLA-4:ipilimumab complex. (A) The concave CTLA-4 front face, formed by the CC′ (beige) and FG (MYPPPY loop; dark purple) strands, is buried between CDRs of ipilimumab. CDRs, FG, and CC′ strands are colored: LCDR3 (olive) and LCDR1 (teal) make the hydrogen bonding interaction with the G strand. HCDR2 (pink) and HCDR1 (light gray) pack against F and C strands and participate in both hydrogen bond and hydrophobic interactions. HCDR3 (magenta) inserts into the center of the front face of CTLA-4 and exclusively participates in hydrophobic interaction. (B) Rainbow representation (N to C termini transition from blue to red) of CTLA-4 molecules with front and back strands labeled white and black, respectively. The G-strand β bulge conserved among the antigen receptors is shown in red. (C) CTLA-4 rotated 90° around the vertical axis relative to B, highlighting the concave front surface, with approximate location of CDRs represented by arrows in respective color (as used in A). The coordinates of the ipilimumab:CTLA-4 complex have been deposited in PDB as ID code 5TRU.

View this table:
Table 1.

Crystallographic data and refinement statistics

Fig. S1.
Fig. S1.

Structure of CTLA-4:ipilimumab (Fab) complex. (A) Two independent copies of CTLA-4:ipilimumab (Fab) complex found in the asymmetric unit. Both CTLA-4 molecules in the asymmetric unit are colored yellow. Heavy (H) and light (L) chains are labeled with colors similar to Fig. 1 (PDB ID code 5TRU). (B) Orientation of the complex showing the concave surface of CTLA-4 and the complementarity of the CDRs that fit the concave surface formed by the FG loop on one side (purple) and the CC′ loop on the other side (beige) of the CTLA-4 IgV domain.

The binding interface is formed by residues from the C, C′, F, and G strands (Fig. 1 B and C) of the front β-sheet of CTLA-4 and light chain complementarity determining regions 1 and 3 (LCDR1 and LCDR3) and heavy chain CDRs 1, 2, and 3 (HCDR1, HCDR2, and HCDR3) (Fig. 1). The only potential interaction involving LCDR2 is the 4.2 Å approach between the side-chain hydroxyl of 50Tyr and the main-chain carbonyl of 44Ser on CTLA-4 [numbering is consistent with previous CTLA-4 structural reports; e.g., Protein Data Bank (PDB) ID codes 1I85 and 1I8L (29, 30)]. CTLA-4 residues from the F and G strands (93Ile, 95Lys, 97Glu, 106Leu, and 108Ile), the FG loop (99MYPPPY104) as well as 33Glu, 35Arg, and 39Leu from the C strand and 46Val from the C′ strand form an extended interface with residues from HCDR2 (52Ser, 53Tyr, 57Asn, and 59Tyr), HCDR3 (101Trp and 102Leu), LCDR1 (31Ser and 33Tyr), and LCDR3 (93Gly, 94Ser, 95Ser, and 97Trp) (Fig. 2A). The G strand from CTLA-4 is positioned along the cleft formed between the VL (LCDR1 and LCDR3) and VH (HCDR1 and HCDR2) domains (Fig. 2 A and B), whereas the adjacent F strand packs against HCDR1 and HCDR2 (Fig. 2A). Together these contacts account for almost all of the hydrogen bonding interactions within the complex. 101Trp and 102Leu from HCDR3 contact the center of the concave hydrophobic patch on the front face of CTLA-4 formed by 39Leu, 46Val, and 93Ile, which extends the binding interface toward the CC′ loop located opposite the FG loop (Fig. 2A).

Fig. 2.
Fig. 2.

Critical interactions between ipilimumab and CTLA-4. (A) Surface representation of ipilimumab (heavy chain in light gray; light chain in dark gray) showing the cleft formed between the VH and VL domains. Residues on the FG (purple) and CC′ (beige) strands of CTLA-4 facing toward the Fab are shown in stick representation. For clarity, side chains on the FG and CC′ strands facing away from the Fab are not shown. The G strand extends along the cleft with hydrophobic residues (104Tyr–108Ile) intercalating into the cleft between the two domains. The F and C strands stack against HCDR2 and HCDR1. 101Trp and 102Leu from the tip of the HCDR3 loop contacts the center of the CTLA-4 front face and interacts with a hydrophobic patch formed by 39Leu, 46Val, and 93Ile. (B) Stick representation of the 99MYPPPY104 loop followed by G-strand residues (purple). The main-chain carbonyl and amide groups of the edge G-strand residues not involved in interstrand interactions with the F strand participate in hydrogen bonding interactions with LCDR3 (olive) and LCDR1 (teal) residues. The 99MYPPPY104 loop (FG loop) packs against a cluster of aromatic residues, and 59Y(pink) from HCDR2 contacts the center of the FG loop and also participates in a hydrogen bonding interaction with the carbonyl oxygen of 99M.

Overall, ∼13 hydrogen bonds and more than 90 contacts less than 4.0 Å contribute to the CTLA-4:ipilimumab Fab interface (Table S1). The total surface area buried at the binding interface is ∼1,880 Å2, with 990 Å2 contributed by the ipilimumab Fab and 890 Å2 by CTLA-4, which is at the high end of observed values (i.e., 1,175–1,755 Å2) for antigen–antibody complexes (31). These extensive interactions are consistent with the high affinity interaction between CTLA-4 and the ipilimumab Fab [equilibrium dissociation constant (Kd) of 10.6 nM]. Both the heavy chain (605 Å2) and light chain (385 Å2) make significant contributions to the binding interface. The F and G strands, which include the 99MYPPPY104 loop, bury 498 Å2 of surface area, representing ∼60% of the total buried surface area (890 Å2) contributed by CTLA-4. The 100YPPP103 segment buries only 56 Å2 of surface area upon binding ipilimumab, with the major contributors from the CTLA-4 99MYPPPY104 loop being Tyr-104 and Met-99, which bury 115.0 and 85.3 Å2 of surface area, respectively. Thus, although the FG loop is involved in the recognition of ipilimumab, this loop is not completely buried, as observed in the CTLA-4:B7-1 and CTLA-4:B7-2 complexes (29, 30).

View this table:
Table S1.

Contacts less than 4.0 Å observed between Ipilimumab and CTLA-4

An important consideration is that the species crystallized in these studies is monomeric CTLA-4, whereas the physiologically relevant cell surface receptor is a disulfide-linked homodimer involving Cys-122, which is outside the well-ordered globular domain and not included in the construct used in this crystallographic analysis. Structural and modeling analyses demonstrate that ipilimumab could readily accommodate the observed interactions within the bona fide CTLA-4 dimer. In particular, the unique mode of CTLA-4 dimerization, involving “side-to-side” contact between CLTA-4 monomers, places the FG loops distal to the dimer interface, such that the relevant epitope in each monomer is highly solvent accessible and appropriately positioned to recapitulate the interactions observed between the CTLA-4 monomer and the intact ipilimumab mAb (Fig. S2).

Fig. S2.
Fig. S2.

Ipilimumab:CTLA-4 complex superposed onto the physiologically relevant CTLA-4 dimer (PDB ID code 3BX7): The front faces of the two CTLA-4 molecules in the dimer that contact ipilimumab are positioned distal to the CTLA-4 dimer interface, consistent with in vivo recognition of CTLA-4 by ipilimumab.

Cell-Based Evaluation of the CTLA-4:Ipilimumab Recognition Interface.

To confirm the crystallographically observed binding interface, we generated a library of 118 dimeric CTLA-4 mutants, which were transiently expressed in HEK293 cells and challenged with soluble ipilimumab. This approach provides a gold standard for evaluating extracellular interactions, as the members of the mutant library undergo all requisite co- and posttranslational modifications (e.g., correct disulfide bond formation and glycosylation) and are presented in the context of the mammalian cell surface environment (32). In the context of cell surface expression, mutation of residues S20, R35, R40, Q76, D88, K95, E97, Y104, L106, and I108 of human CTLA-4 resulted in significant loss of ipilimumab binding (Fig. 3 A and B and Fig. S3A). Of the residues identified—R35, K95, E97, Y104, L106, and I108—are observed to make direct contacts with ipilimumab in the crystal structure. Residues R40 and D88 form a salt bridge that links two loops in the membrane proximal region of CTLA-4; disruption of this interaction likely causes local or global structural perturbations, resulting in loss of ipilimumab binding. Similarly, mutation of S20 or Q76 to aspartic acid also resulted in loss of ipilimumab binding, despite residing on the opposite side of CTLA-4 relative to the ligand-binding site. O- and N-linked glycosylation algorithms predict that S20 and N75 (adjacent to Q76) are potentially glycosylated, suggesting that mutations at residues 20 and 76 may affect installation or interaction with the carbohydrate moieties and potentially overall structural stability (Fig. S3B) (33). The likelihood that mutations at these four residues (S20, R40, N75, and D88) cause global structural perturbation/destabilization of CTLA-4 is supported by the observation that they also result in the loss of binding to B7-1, B7-2, and ICOS-L, despite the observation that these residues reside outside of the experimentally determined binding interfaces (Fig. 3 A and B and Fig. S3C).

Fig. 3.
Fig. 3.

Mapping the ipilimumab-binding epitope on cell surface-expressed human CTLA-4. (A) Wild-type and mutant CTLA-4 constructs (118 total) were transiently expressed as C-terminal mCherry fusions in HEK293 suspension cells. Two days posttransfection, CTLA-4–expressing cells were queried with either the ipilimumab monoclonal antibody or cells transiently expressing hB7-1-GFP, hB7-2-GFP, or hICOS-l-GFP and analyzed by flow cytometry to determine the percentage of mCherry-positive cells (CTLA-4–expressing) bound (anti-human Alexa 488 signal reports on ipilimumab binding; GFP signal reports on B7-1, B7-2, or ICOS-L binding). The chart highlights (colored background) the average percent bound and SD from three independent experiments for those CTLA-4 mutants that resulted in ≤50% binding to a particular query. (B) Residues at which mutations caused a significant loss of ipilimumab binding are mapped onto the CTLA-4 crystal structure. The residues highlighted in red make direct contact with ipilimumab, the two blue residues R40 and D88 form a salt bridge, and the two green residues (S20 and Q76) affected ipilimumab binding even though they are distal to the ipilimumab recognition surface.

Fig. S3.
Fig. S3.

FACS-based interaction mapping. (A) Data showing the average percent bound of three independent epitope mapping experiments for ipilimumab, B7-1, B7-2, and ICOS-L to CTLA-4 mutant library, with error bars showing SD. (B) Residues on hCTLA-4 predicted to be O- or N-glycosylated. Residues in red were predicted using two different prediction algorithms, whereas those in black were predicted either by Hamby et al. (33) or NetOGly 4.0 (bold) but not both. (C) Crystal structure of the hB7-1:hCTLA-4 complex (PDB ID code 1I8L) and the hB7-2:hCTLA-4 complex (PDB ID code 1I85) highlighting residues on CTLA-4 that affected binding of B7-1 or B7-2 when mutated. The residues highlighted red showed <10% binding to ligand when mutated, and tan residues showed 20–50% binding when mutated. The two blue residues (R40 and D88) form a salt bridge, and the two green residues (S20 and Q76) also significantly affected ipilimumab binding. (D) Structure of the hCTLA-4 homodimer (PDB ID code 3OSK) (Left) and top view x-rotated 90° (Right). The residues highlighted green show the location of the MYPPPY loop involved in ipilimumab, B7-1, B7-2, and ICOS-L binding. The residues highlighted as yellow sticks show those residues, which although distal to the binding surface, significantly affected B7-1, B7-2, and ICOS-L binding when mutated. (E) Data show the percentage of gated mCherry-positive cells (CTLA-4 expression) normalized to wild-type CTLA-4 mCherry expression (far right). The error bars show the SD in transient expression from three independent transfections.

These mapping studies also highlight features of the binding surface recognized by B7-1, B7-2, and ICOS-L. A number of CTLA-4 mutations (R35A, R35D, K95A, K95D, E97A, E97R, Y104A, Y104D, I108A, and I108D) that strongly impair B7-1 and B7-2 binding are consistent with contacts observed in the CTLA-4:B7-1 and CTLA-4:B7-2 crystal structures (29, 30) (Fig. S3C). In addition, mutations at residues E33, E48, and Y100, also present at the CTLA-4:B7-2 binding interface, result in modest reductions in binding between CTLA-4 and B7-2. Again, mutations at residues R35, K95, E97, Y104, and I108 severely diminished ipilimumab binding, consistent with the direct steric blockade of CTLA-4 ligand binding by ipilimumab. This conclusion is further supported by results from a competition experiment demonstrating that ipilimumab, but not a control mAb, directly competes with hB7-1 for binding to beads coated with hCTLA-4 (Fig. 4 A–C). In addition, mutation of a number of CTLA-4 residues distal to the ligand-binding interfaces (e.g., V10, L12, S14, T69, T71, N78, Q82, G83, R85, A86, Y115, D118, and E120) exhibited impaired binding to B7-1, B7-2, and ICOS-L (Fig. S3D). Importantly, all of the CTLA-4 mutants analyzed for binding showed similar mCherry expression (Fig. S3E), making it unlikely that these mutations caused severe structural instability or protein misfolding. It is more likely that mutations of residues on CTLA-4 distal to the observed ligand binding site caused more subtle local structural perturbations. Three of these residues (Y115, D118, and E120) contribute to the CTLA-4 dimer interface, with the remaining residues arranged in a patch that extends toward the ligand binding surface (29, 30). The mechanistic underpinnings for the binding properties of these mutants is not clear but may be related to the previous report that CTLA-4 lacking the interchain disulfide exists as an interconverting population of noncovalent dimer and monomer, which fully resolves to monomer when bound to B7-2 (34). These observations suggest a mechanism that couples ligand binding with formation of the CTLA-4 dimer interface. The biological consequences of this behavior will require further evaluation.

Fig. 4.
Fig. 4.

Ipilimumab directly competes with hB7-1 for binding to hCTLA-4 in a bead-based FACS assay. (A) Schematic representation of the assay. (B) Protein A beads saturated with hCTLA-4 hIgG1 were titrated with hB7-1 mIgG2a protein. FACS analysis was used to determine the GeoMean of the FL1 channel (488) for ALL BEADS gated. Data represent the average of three independent experiments with error bars showing the SD. (C) Protein A beads loaded with hCTLA-4 hIgG1 were saturated with hB7-1 mIgG2a (5 nM) and incubated with increasing concentrations of control mAb or ipilimumab. Data show the GeoMean of Fl1 normalized to the 5-nM titration point from that same experiment. All data show an average of three independent experiments with error bars showing SD.

Biochemical Confirmation.

To further evaluate the contributions to the binding interface, a selected subset of CTLA-4 mutants were purified and their interactions with ipilimumab Fab were evaluated by size exclusion chromatography (SEC) and native gel analysis. In particular, 95Lys, 99Met, and 104Tyr, which all contact the Fab, and 105Tyr, which lies outside the crystallographically observed binding interface, were mutated to alanine in dimeric human CTLA-4 containing the native interchain disulfide. Interactions between the wild-type and mutant dimeric CTLA-4 molecules and the Fab were examined by SEC (Figs. 5A and 6A). The dimeric K95A and Y104A mutants did not exhibit any interaction with the Fab, whereas the dimeric M99A mutant exhibited significantly decreased association compared with the wild type (Fig. 6A). The Y105A mutation did not significantly affect the interaction between dimeric CTLA-4 and the Fab, consistent with the placement of 105Y outside the observed ipilimumab recognition interface (Fig. 6A). The behavior of the wild-type dimeric CTLA-4 is consistent with the function blocking activity of intact ipilimumab.

Fig. 5.
Fig. 5.

Biochemical confirmation of the ipilimumab:CTLA-4 binding interface, and formation of the wild-type dimeric hCTLA-4:Fab complex demonstrated by gel filtration and native PAGE gel. (A) Gel filtration of dimeric hCTLA-4 and Fab complex. The chromatograms for Fab, dimeric hCTLA-4, and the dimeric hCTLA-4:Fab mixture are indicated as black, blue, and red curves, respectively. It should be noted that the ipilimumab Fab (molecular mass, ∼50 kDa) reproducibly exhibits aberrantly long retention times relative to the smaller dimeric CTLA-4 (molecular mass, ∼25 kDa). (B) Native PAGE analysis of the dimeric hCTLA-4 and Fab complex. The concentration of wild-type dimeric hCTLA-4 was held constant at 10 μM with increasing concentration of Fab. Wild-type dimeric hCTLA-4 and Fab were mixed in molar ratios of 4:1, 2:1, 1:1, 1:2, and 1:4 from lane 1 to lane 5. Lanes 6 and 7 show Fab and dimeric hCTLA-4 alone, respectively.

Fig. 6.
Fig. 6.

Gel filtration and native PAGE analysis of dimeric mutants and Fab. (A) The chromatograms for Fab, dimeric hCTLA-4 mutants, and the dimeric hCTLA-4:Fab mixture are indicated as black, blue, and red curves, respectively. (Top Left) K95A. (Top Right) Y104A. (Bottom Left) M99A. (Bottom Right) Y105A. All peaks of the corresponding proteins are indicated by arrows. The formation of the complex is indicated by a red arrow. (B) Distinct migration bands for dimeric hCTLA-4 mutants, Fab, and the hCTLA-4 mutant:Fab complex are diagnostic for interactions of dimeric hCTLA-4 mutants and the Fab. The concentrations of dimeric hCTLA-4 mutants were held constant at 10 µM. (Top Left) Dimeric mutant K95A and Fab were mixed in molar ratios of 4:1, 2:1, 1:1, 1:2, and 1:4, from lane 1 to lane 5. Lanes 6 and 7 show Fab and dimeric mutant K95A alone, respectively. (Top Right) Dimeric mutant Y104A and Fab were mixed in molar ratios of 4:1, 2:1, 1:1, 1:2, and 1:4, from lane 1 to lane 5. Lanes 6 and 7 show Fab and dimeric mutant Y104A alone, respectively. (Bottom Left) Dimeric mutant M99A and Fab were mixed in molar ratios of 2:1, 1:1, and 1:2, from lane 1 to lane 3. Lanes 4 and 5 show Fab and dimeric mutant M99A alone, respectively. (Bottom Right) Dimeric mutant Y105A and Fab were mixed in molar ratios of 2:1, 1:1, and 1:2, from lane 1 to lane 3. Lanes 4 and 5 show Fab and dimeric mutant Y105A alone, respectively.

The interactions between the Fab and the wild-type dimeric CTLA-4 and dimeric mutants were further evaluated by native nonreducing PAGE analysis. The ipilimumab Fab does not enter the gel under native PAGE conditions due to the overall positive charge of the Fab (pI of 8.8) at pH 8.0. Formation of the dimeric CTLA-4:Fab complex results in a species capable of entering the gel, whereas unbound dimeric CTLA-4 results in a distinct band (Fig. 5B). As shown by the native PAGE results, the wild-type dimeric CTLA-4 interacted with Fab and formed new species consistent with the formation of the complex in the gel (Fig. 5B). Dimeric K95A, M99A, and Y104A CTLA-4 mutants did not form any new species with ipilimumab Fab under native PAGE conditions (Fig. 6B). Dimeric Y105A CTLA-4 productively interacted with the Fab, as demonstrated by the appearance of a new band in the native PAGE (Fig. 6B). Importantly, as evidenced by fluorescence-monitored thermal denaturation, all of these dimeric CTLA-4 mutants are fully folded under the conditions used in these experiments (SI Materials and Methods and Fig. S4). These results are fully consistent with the observed crystallographic interface and support the physiological relevance of the CTLA-4: ipilimumab Fab complex. Interactions of B7 ligands with wild-type CTLA-4 and mutants were confirmed by native nonreducing PAGE analysis as well as by SEC (Figs. S5, S6, and S7).

Fig. S4.
Fig. S4.

Thermal melting of wild-type dimeric hCTLA-4 and dimeric mutants. (A) The Sypro-orange fluorescence is plotted against the temperature. (B) The midpoint of the unfolding transition defines the melting temperature of the corresponding protein. The melting temperature of each mutant and wild-type hCTLA-4 is the average value of two parallel groups. All mutants exhibited thermal denaturation properties similar to the wild-type dimeric CTLA-4 (Tm = 54 °C). These data indicate that the mutants were correctly refolded under the conditions used for the binding experiments.

Fig. S5.
Fig. S5.

Formation of wild-type dimeric hCTLA-4:hB7-2 complex is indicated by gel filtration and native PAGE gel. (A) Gel filtration of dimeric hCTLA-4 and monomeric hB7-2 complex. The chromatograms for hB7-2, dimeric hCTLA-4, and the dimeric hCTLA-4:monomeric hB7-2 mixture in a molar ratio of 1:2 are indicated as black, blue, and red curves, respectively. (B) Native PAGE analysis of the dimeric hCTLA-4 and monomeric hB7-2 complex. The concentration of wild-type dimeric hCTLA-4 was 10 μM. Wild-type dimeric hCTLA-4 and monomeric hB7-2 were mixed in molar ratios of 4:1, 2:1, 1:1, 1:2, and 1:4, from lane 1 to lane 5. Lanes 6 and 7 are hB7-2 and hCTLA-4 alone, respectively.

Fig. S6.
Fig. S6.

Gel filtration and SDS/PAGE analysis of dimeric mutant K95A and dimeric wild-type hCTLA-4. (A) Overlay of Superdex 200 gel filtration of dimeric mutant K95A (blue solid) and wild-type dimeric hCTLA-4 (red solid). (B) SDS/PAGE analysis of dimeric mutant K95A with DTT (lane 1) or without DTT (lane 2) and wild-type dimeric hCTLA-4 with DTT (lane 3) or without DTT (lane 4).

Fig. S7.
Fig. S7.

Gel filtration analysis of dimeric mutant Y105A with monomeric hB7-1 or hB7-2. The chromatograms for monomeric hB7-1 or hB7-2, dimeric hCTLA-4 mutant Y105A, and the dimeric mutant Y105A:monomeric hB7-1 or hB7-2 complexes are indicated as black, blue, and red curves, respectively. (A) Overlay of Superdex 200 gel filtration of dimeric mutant Y105A (blue solid), monomeric hB7-1 (black solid), and the dimeric mutant Y105A:monomeric hB7-1 mixture. (B) Overlay of Superdex 200 gel filtration of dimeric mutant Y105A (blue solid), monomeric hB7-2 (black solid), and the dimeric mutant Y105A:monomeric hB7-2 mixture.

Epitope Verification and Specificity Determinants of Ipilimumab.

The challenge of intact ipilimumab with proteolytically generated (i.e., AspN, GluC, and Trypsin) peptide fragments derived from the disulfide-linked human CTLA-4–Fc fusion protein, coupled with MS, identified a discontinuous epitope involving three different linear peptides (Fig. S8). P1 stretches from 26YASPGKATEVRVTVLRQA42, P2 from 43DSQVTEVCAATYMMGNELTFLDD65, and P3 from 96VELMYPPPYYLGIG109. From Table S1, it can be seen that residues from HCDR2 and HCDR3 interact with P1 and P2, respectively, whereas HCDR2, HCDR3, and LCDR3 interact with P3. Most of the interaction energy between CTLA-4:B7-1 and CTLA-4:B7-2 is derived from contacts with 99MYPPPY104 (9), which is contained within the P3 segment recognized by ipilimumab, consistent with the function blocking activity of ipilimumab.

Fig. S8.
Fig. S8.

Several peptides of CTLA-4 produced by reduction, alkylation, and digestion by trypsin protease were bound to ipilimumab, eluted, and detected by LC–MS/MS. Of these, the stretch 98LMYPPPYYLGIG109 bound to ipilimumab. The corresponding stretch in CD28, 96IEVMYPPPYLDNEK109, did not bind to ipilimumab, although it contains the 99MYPPPY104 sequence that is important for binding both B7-1 and B7-2.

To understand the specificity of ipilimumab toward CTLA-4, linear peptides from both native and reduced human CD28–Fc fusion protein were generated, and their interactions with ipilimumab-coupled beads were examined. After elution from antibody-decorated beads, the mass and sequences of the bound peptides were characterized by MALDI-TOF MS and nano-liquid chromatography (nano-LC)–MS/MS, respectively. Of particular note is the CD28 tryptic peptide, 96IEVMYPPPYLDNEK108, which did not bind ipilimumab (Fig. S8). In contrast, the corresponding peptide from CTLA-4, 98LMYPPPYYLGIGN110, which is part of peptide P3 (Fig. S8), bound well to ipilimumab. Even though the sequence 99MYPPPY104 was present in both linear peptides from CTLA-4 and CD28, the observation that only the CTLA-4–derived peptide binds implies that residues outside of this region play important roles in recognition of the 99MYPPPY104-containing epitope, which is consistent with the crystallographic data as discussed above (also see Table S1).

Mechanism of CTLA-4 Recognition and Blockade by Ipilimumab.

The ability of CTLA-4 and CD28 to recognize both the B7-1 and B7-2 ligands is, in part, the consequence of the conserved FG loop (99MYPPPY104) shared by both receptors (Fig. 7). The introduction of mutations in this loop resulted in greater than 90% loss of affinity to the B7 ligands, identifying this segment as the core of the ligand binding surface on both CTLA-4 and CD28 (9). Direct structural analyses of the CTLA-4:B7-1 and CTLA-4:B7-2 complexes (29, 30) showed that the 99MYPPPY104 loop contributes ∼80% of the interfacial contacts with the B7 ligands. The three consecutive proline residues in the FG loop assume an unusual cis–trans–cis conformation, and with the exception of 99Met, all of the main-chain carbonyl atoms are directed away from the ligand binding surface (Fig. 8A). Given its chemical composition and conformation, this loop is devoid of free amide nitrogens or solvent-accessible carbonyls (except for the terminal residues 99M and 104Y) and thus lacks determinants typically associated with directionality and specificity (i.e., hydrogen bond donors and acceptors). This unique and highly strained main-chain conformation provides significant geometric complementarity and hydrophobicity to support recognition of the concave surfaces presented by the front sheets of the B7 ligands. The structures of the unliganded murine and human CTLA-4 molecules demonstrate that this FG loop adopts essentially the same detailed conformation in both forms (35), indicating that CTLA-4 presents a preformed B7-recognition surface. Although a structure of CD28 bound to a B-7 ligand is not available, the conserved sequences and conformation of the FG loop in the structure of a CD28:Fab complex (36), as well a series of mutagenesis experiments (9), supports a mode of interaction with B7 ligands similar to that observed for CTLA-4.

Fig. 7.
Fig. 7.

Sequence alignment of CTLA-4 and CD28. CTLA-4 residues interacting with ipilimumab are marked with blue triangles. The filled circles indicate residues from B7-1 (crimson) and B7-2 (purple) that make contacts less than 4.0 Å. Insertion in CD28, before the β bulge on the G strand, is shown with an arrow. Notice the sequence differences in the G-strand residues just after the 99MYPPPY104 loop. The three discontinuous stretches of CTLA4 (P1, 26YASPGKATEVRVTVLRQA42; P2, 43DSQVTEVCAATYMMGNELTFLDD65; and P3, 96VELMYPPPYYLGIG109) identified as part of the ipilimumab epitope, by nano-LC–MS/MS, are shown by brown lines above the sequence of CTLA-4.

Fig. 8.
Fig. 8.

Mechanism of ipilimumab action. (A) Identical residues between CTLA-4 and CD28 are colored blue, and insertions in the CTLA-4 sequence are colored red; both are mapped onto the CTLA-4 structure (yellow). (B) Identical residues between CTLA-4 and CD28 are colored blue, and insertions in the CD28 sequence are colored red; both are mapped onto the CD28 structure (yellow). Note that the MYPPPY-loop surfaces contours are similar. (C) Mode of B7-2 (light blue) interaction with the MYPPPY-loop surface. (D) Mode of B7-1 (pink) interaction with the MYPPPY-loop surface. (E) Mode of CTLA–4 and ipilimumab interactions. (F) Superposition of the CTLA-4:B7-2 and CTLA-4:ipilimumab structures, based on the CTLA-4 component in each complex, is presented. Superposition of the CTLA-4:B7-1 and CTLA-4:ipilimumab complexes is shown in Fig. S9. These superpositions indicate that ipilimumab and the B7 ligands compete for overlapping binding surfaces on CTLA-4.

Ipilimumab exploits the unique features of the FG loop and sandwiches the front sheet of CTLA-4 between LCDR1 and LCDR3 on one side and HCDR1 and HCDR2 on the other (Fig. 2A). The region between LCDR3 and HCDR2 is rich in aromatic residues and stacks against the surface presented by the 99MYPPPY104 loop (Fig. 2B). Residues in the tips of LCDR3 and LCDR1 are positioned proximal to the edge G-strand side of the front sheet of CTLA-4 and participate in a series of hydrogen bonding interactions with main-chain atoms of the G strand (Fig. 2B). HCDR1 and HCDR2 stack almost perpendicular to the C, F, and G stands of CTLA-4 and provide both hydrophobic and hydrogen bonding interactions with the solvent-exposed residues on the F and C strands as well as those located toward the termini of the FG loop (Table S1). The HCDR3 loop contacts the center of the concave surface of CTLA-4 on its front sheet, with interactions extending toward the CC′ loop (Figs. 1A and 2A). In the CTLA-4:B7 complexes, the tip of the FG loop (99MYPPPY104) from CTLA-4 contacts the front face of the B7 ligands (Fig. 8 C and D and Fig. S9 A and B), with the CC′ loop of CTLA-4 positioned away from the B7 surface. However, in the CTLA-4:ipilimumab complex, the –101PPP103− tip of the FG loop is only partially engaged, allowing for the association of the entire CTLA-4 front face with ipilimumab (Fig. 8E and Fig. S9C). In this way, the ipilimumab interaction surface not only sterically occludes the conserved 99MYPPPY104 surface from availability to B7 ligands but also extends its interaction toward the opposite side of the CTLA-4 IgV domain (i.e., CC′ loop) (Fig. 8F). The recognition of this extended CTLA-4 surface by ipilimumab (relative to the B7 ligands) is the consequence of the highly twisted architecture common to the front sheet of IgV domains of the antigen receptors, which results in the protrusion of the long FG and CC′ loops away from the plane of the front sheet, creating a concave surface on the CTLA-4 front sheet (Fig. 1C) (35).

Fig. S9.
Fig. S9.

Mode of B7-1 (magenta), B7-2 (orange), and ipilimumab [green (H) and red (L)] interaction with CTLA-4 (yellow). B7-1 and B7-2 ligands interact with CTLA-4 using their front face closer to the tip of the CTLA-4 FG-loop (purple) (A and B). Ipilimumab interaction is relatively shifted toward the center of the front face of CTLA-4 (C).

The differences in total buried surface area upon binding CTLA-4 (∼1,885 Å2 for ipilimumab vs. ∼1,250 Å2 for the B7 ligands), the greater number of hydrogen bonds (∼13 for ipilimumab vs. ∼5–7 for the B7 ligands), and the larger number of van der Waals contacts are consistent with the Kds exhibited by ipilimumab and B7 ligands for CTLA-4 (∼10 nM and ∼0.1–1 μM range, respectively) and the observation that ipilimumab competes effectively with the B7 ligands for binding CTLA-4. It is this direct competition between ipilimumab and the B-7 ligands that, in part, underlies the therapeutic efficacy of ipilimumab.

Ipilimumab Specificity.

Although the B7 ligands interact with both CTLA-4 and CD28, the ability of ipilimumab to discriminate between these two functionally distinct receptors is critical for its therapeutic efficacy. CTLA-4 and CD28 share ∼30% sequence identity, including conservation of the critical core residues and disulfide linkages that are essential for maintaining the IgV domain fold. Overall, the structure of the Fab-bound human CTLA-4 monomer is similar to that in the apo–CTLA-4 homodimer structure (35), the CTLA-4:B7-1 complex (30), the CTLA-4:B7-2 complex (29), and the CTLA-4:Lipocalin (37) complex, with Cα–rmsds that range between 0.85 and 1.1 Å. Similarly, the Cα–rmsds calculated with ipilimumab-bound human CTLA-4 against human CD28 (36) and murine CTLA-4 (38) structures are 1.51 Å and 1.27 Å, respectively.

Nearly all features of the FG loop (99MYPPPY104) are conserved among the structurally characterized CD28:CTLA-4 family members, including the sequences and the specific side-chain conformations; the sole exception is found in the apo–CTLA-4 homodimeric structure, where 100Tyr adopts a different rotamer conformation to accommodate crystal contacts (35). The conformation of this loop in the Fab-bound CD28 structure is also similar to that found in CTLA-4, including side-chain rotamers, with an overall Cα–rmsd of ∼0.3 Å (Fig. 8 A and B). Based on these observations, these segments are not likely to be major determinants for the discrimination between CTLA-4 and CD28. Although many other CTLA-4 residues that interact with ipilimumab are conserved in CD28, there are a few important differences in sequence and conformation. 39Leu and 93Ile in CTLA-4, which interact with 101Trp and 102Leu from HCDR3, are replaced in CD28 by 38His and 93Phe, respectively, and may contribute to the ability of ipilimumab to discriminate between CTLA-4 and CD28 (Fig. 8).

As members of the antigen receptor group of the IgSF (35), both CTLA-4 and CD28 possess β bulges in C′ and G strands, which in CTLA-4 are centered on 48Glu and 110Asn, respectively (in CD28, these bulges are centered on 46Glu and 110Asn, respectively) (Fig. 1B) (39). The –Gly–X–Gly– sequence of G-strand β bulge observed in CTLA-4 is present in all VL domains and more than 98% of VH domains of mABs (39). Notably, in CD28, the first glycine in this sequence is replaced by serine. In CD28, the single residue insertion 108Glu (between the equivalent of 108Gly and 109Ile in CTLA-4), just before the G-strand β bulge, exaggerates the protrusion of the G strand (Figs. 1B, 7, and 8B). In addition, on the G strand, immediately following the conserved 99MYPPPY104 loop, all residues within the span of 105–109 are different in CD28 (Figs. 7 and 9). In particular, 107Asn preceding the insertion is involved in two hydrogen bond interactions with the F-strand backbone, which further perturbs the canonical interstrand interaction between the F and G strands (Fig. 9 B and C), pushing the two strands farther apart than is typical. In CTLA-4, this region is involved in continuous main chain–main chain and main chain–side chain interactions with LCDR1 and LCDR3 (Fig. 2B). Superposition of CD28 on the CTLA-4:ipilimumab complex predicts that the distortion of the G strand results in not only loss of hydrogen bonds but also a substantial steric clash with LCDR1 (Fig. 9C). The detailed structural differences in the G strands of CTLA-4 and CD28 are likely the major determinants responsible for the specificity exhibited by ipilimumab. In addition, in CTLA-4 there is an insertion of residue 47T just before the C′-strand β bulge (between the equivalent of residues 45V and 46E in CD28; Figs. 7 and 8A), which probably contributes to the differences in the CC′ loop conformation. However, this insertion is distal to the recognition interface and is not predicted to significantly impact the binding or selectivity.

Fig. 9.
Fig. 9.

Determinants of selectivity. (A) The 99MYPPPY104 loop conformations in CTLA-4 (purple) and CD28 (yellow) are very similar. Note that only the carbonyl moiety from 99M and the amide nitrogen from 104Y are facing toward the viewer and are available for interaction with the ligands. (B) Insertion in the edge G strand following the conserved β bulge causes the G strand of CD28 to protrude away from the F strand, resulting in the loss of many interstrand interactions with the F strand. (C) Possible clashes between the LCDR1 and the extended protrusion of the G strand due to insertion at 109K on CD28 (yellow). The G strand from CTLA-4 is colored purple.

An important consideration for normal physiology and therapeutic efficacy is that an optimal biological outcome is not always associated with the strongest achievable binding (i.e., very high affinity between receptor and ligand). This concept is illustrated by numerous physiological processes, such as the need for reversible binding interactions in oxygen delivery, replication, transcription, and translation, and these same principles are directly relevant to the development of therapeutic strategies (reviewed in ref. 40). Notable examples are provided by recent reports of chimeric antigen receptor (CAR) T cells engineered to present scFv modules spanning a range of affinities for the target antigen (41, 42). These studies demonstrate that scFvs with reduced affinities had increased selectivity for solid tumors, relative to normal cells, due to the requirement for higher surface density of target antigens to support productive engagement, a state often afforded by the targeted tumor cells. For example, CAR T cells bearing scFvs derived from nimotuxumab, but not cetuximab (43), effectively discriminated between malignant cells and nonmalignant cells based on its ∼10-fold poorer Kd for EGFR, which is expressed at significantly higher levels on the malignant cells (41). Similar studies, which examined scFvs against EGFR (derived from the C10 anti-EGFR antibody) (44) and ErbB2 (derived from the 4D5 trastuzumab antibody) (45) with Kds spanning ∼2–3 orders of magnitude, have reinforced this concept. Together, these studies demonstrate that apparent Kds not only control selectivity between target and nontarget cells but can also impact CAR T-cell function (e.g., cytokine production) and suggest that on and off rates (i.e., kon and koff) will have important mechanistic contributions to the ultimate therapeutic function. These examples and considerations underscore the need to achieve optimal, not maximal, affinities (and kinetics) to attain the desired biological/therapeutic activity. In the case of ipilimumab, affinity-attenuated variants could be generated, which would exhibit selectivity for activated Treg cells present at tumor sites due to their higher expression levels of CTLA-4, compared with other activated T cells (20, 21). Additionally, it might be possible to engineer ipilimumab variants that are more responsive to the reduced pH often associated with the tumor microenvironment (46⇓–48). Thus, the structure of the CTLA-4:ipilimumab complex provides the foundation for the rationale design of affinity-modulated ipilimumab variants with distinct T-cell subset-targeting profiles and more selective and efficacious therapeutic properties.

Materials and Methods

Detailed methods can be found in SI Materials and Methods.

Monomeric and dimeric CTLA-4 and the monomeric IgV domain of hB7-2 were refolded as previously described from inclusion bodies (29, 49). Ipilimumab Fab fragments were prepared by papain digestion. Crystals of the CTLA-4:ipilimumab Fab complex were obtained by sitting drop vapor diffusion and the structure determined by molecular replacement. The crystallographically determined binding interface was confirmed by a high-throughput FACS analysis using a library of CTLA-4 mutants expressed on the surface of HEK293 cells as well as by direct biochemical characterization of ipilimumab binding to wild-type CTLA-4 and mutant CTLA-4 variants.

SI Materials and Methods

Protein Production.

Expression, purification, and mutagenesis of CTLA-4 and hB7-2.

The expression and refolding of the monomeric (residues 1–118) and homodimeric (residues 1–126, including Cys122 responsible for the interchain disulfide) forms of human CTLA-4 IgV domain were similar to that described for the hB7-2 IgV domain (49). Escherichia coli BL21 (DE3) pLysS was transformed with pET3a (Novagen) containing the appropriate coding sequences and grown in LB medium at 37 °C. Protein expression was induced with 1.0 mM isopropyl 1-thio-d-galactopyranoside when the OD600 reached 0.5. After induction for 4 h, the cells were harvested and resuspended in buffer containing 50 mM Tris·Cl, pH 8.0, 100 mM NaCl, 20% (wt/vol) sucrose, 1 mM EDTA, and 10 mM DTT and lysed by French press.

The resulting inclusion bodies were collected by centrifugation and washed three times with buffer containing 10 mM Tris·Cl, pH 8.0, 10 mM DTT, 100 mM NaCl, 1 mM EDTA, and 0.5% Triton X-100, and then twice more with the same buffer lacking Triton X-100. Approximately 150 mg of inclusion bodies were solubilized in 3 mL of buffer containing 10 mM NaAc, 6 M guanidine hydrochloride (Gu·HCl), 5 mM EDTA, and 1 mM DTT, pH 4.6. A total of 16 mg of solubilized inclusion bodies were diluted into 16 mL of buffer composed of 10 mM NaAc, 6 M Gu·HCl, and 5 mM EDTA, pH 4.6, and rapidly diluted over a few seconds into 1 L of refolding buffer composed of 200 mM Tris·HCl, 0.4 M arginine·HCl, 2 mM EDTA, 5 mM cysteamine, and 0.5 mM cystamine, pH 8.5, at 4 °C with vigorous stirring. Five more aliquots of inclusion bodies (16 mg each) were added every 6–9 h. Refolded protein was concentrated and buffer exchanged into 10 mM Tris·HCl and 20 mM NaCl, pH 8.5, by Amicon ultrafiltration (Millipore) with a 5,000 Da molecular mass cutoff membrane. Refolded protein was purified by SEC on Superdex 200 (GE Healthcare) and analyzed by Superdex G-75 gel filtration columns (GE Healthcare) in buffer composed of 20 mM Hepes, 150 mM NaCl, 1 mM EDTA, and 0.02% Sodium Azide, pH 7.0.

Monomeric and dimeric CTLA-4 and the monomeric IgV domain of hB7-2 exhibited symmetric monodisperse profiles when examined by SEC, with estimated molecular masses of ∼25 kDa for dimeric CTLA-4 and 13 kDa for monomeric hB7-2 (Fig. S5A). The chromatographic behavior of a 1:2 molar mixture of wild-type dimeric CTLA-4 and monomeric hB7-2 IgV domain indicated that the majority of the material formed a noncovalent complex with a 1:2 molar ratio (i.e., each subunit of the CTLA-4 dimer binds a single molecule of hB7-2 IgV domain) (Fig. S5A). Native PAGE results also demonstrated that refolded recombinant wild-type dimeric CTLA-4 and hB7-2 productively interact (Fig. S5B). This behavior is consistent with the previously reported stoichiometry, indicating that the refolded materials used for structural validation in this study adopt normal tertiary and quaternary structures and possess appropriate biochemical activity (29). All refolded dimeric mutants exhibited chromatographic behavior similar to wild-type dimeric CTLA-4, with the exception of dimeric K95A, which unexpectedly eluted somewhat faster than wild-type dimeric CTLA-4 (Fig. S6A). SDS/PAGE analysis with or without DTT suggested that the disulfide bonds of dimeric mutant K95A were correctly formed (Fig. S6B). The gel filtration results are strong evidence that nearly all mutants were correctly refolded and had the potential to interact with the Fab. The dimeric E97A mutant aggregated during refolding, suggesting that 97Glu may play a role in folding and/or stabilization of CTLA-4.

Mutants of dimeric CTLA-4 were generated by PCR-based mutagenesis and purified as described for wild-type CTLA-4. All wild-type and mutant sequences were confirmed by DNA sequencing.

Ipilimumab Fab fragments.

About 100 mg of ipilimumab monoclonal antibody at a concentration of 9 mg/mL in 50 mM sodium phosphate and 10 mM EDTA, pH 7.0, was mixed with 10 mL of immobilized papain resin (Thermo Fisher, Pierce) that was preactivated with 20 mM sodium phosphate, 10 mM EDTA, and 20 mM Cysteine HCl, pH 7.0. The mixture was incubated at 37 °C for 6 h with continuous mixing. Solubilized Fab and Fc fragments as well as undigested IgG1 were recovered by pouring the resin onto an empty column and collecting the flowthrough. The resin was further washed with 50 mL of 0.1 M Tris, pH 8.0, and pooled with the flowthrough fraction. The pool was then dialyzed overnight against 0.1 M Tris, pH 8.0, buffer to remove excess cysteine. The dialyzed material was then mixed with 15 mL of protein A Sepharose resin (GE Healthcare) that was equilibrated with 0.1 M Tris and 0.5 M NaCl, pH 8.0. The mixture was incubated at room temperature for 2 h with continuous mixing. The resin was then poured onto an empty column and the flowthrough fraction containing Fab fragments collected. The column was further washed with 50 mL of 0.1 M Tris and 0.5 M NaCl, pH 8.0, and combined with the flowthrough fraction. This Fab material was dialyzed extensively against PBS, pH 7.4, and concentrated using VivaSpin 20 concentrators (Satorius). Protein concentration was determined by measuring the absorbance of the sample at 280 nM using a UV spectrophotometer.

Purity of the Fab fragment was determined by SDS/PAGE. About 15 μg of ipilimumab Fab fragment was analyzed by SDS/PAGE using 4–20% Tris·Glycine gel under both reducing and nonreducing conditions. SeeBlue Plus 2 markers from Invitrogen were used as molecular mass markers, and the gel was stained with SimplyBlue stain (Invitrogen) for 1 h at room temperature and destained by washing the gel with deionized water overnight. Under nonreducing and reducing conditions, ∼50 kDa and 25 kDa protein species were detected, respectively. LC–MS analysis of Fab fragment was performed to evaluate the site of cleavage by papain, and the Fab preparation was found to contain a single species with a mass of 47,637 Da.

Crystallization and Data Collection.

Complexes of monomeric (residues 1–118) and dimeric (residues 1–126) human CTLA-4:Fab were purified by SEC on Superdex 200 equilibrated with 20 mM Hepes, pH 7.0. Diffraction quality crystals of the complex of monomeric CTLA-4:Fab were obtained by sitting drop vapor diffusion at 17 °C by mixing 1 μL of protein (10 mg/mL) with 1 μL of reservoir solution (0.1 M ammonium sulfate, 0.1 M Tris, pH 7.5, 20% PEG1500). Thin needle-shaped crystals appeared after 2 d, which were briefly transferred to reservoir solution supplemented with 20% PEG400 and flash-cooled in liquid nitrogen before data collection. Data were collected using the micro-beam facility at the 24-ID-E beam line, APS, Argonne National Laboratory. Although measured intensities were weak, diffraction extended to 3.0 Å resolution. Three wedges of consecutive datasets, each covering 60°, were collected from different parts of the same crystal and merged. Data were integrated and scaled with HKL2000 (50). Crystals exhibited diffraction consistent with the space group C2221 (a = 95.72 Å, b = 197.42 Å, c = 148.06 Å).

Structure Determination.

The structure of the monomeric CTLA:4-Fab complex was determined by molecular replacement using the program MOLREP in CCP4 (51). The search model consisted of human CTLA-4 (1I8L) and the chainsaw (CCP4, 1994) truncated model of a Fab fragment (2V7N; IgV domains of H and L chains and IgC domains of H and L chains were used in different searches). The model was improved by alternate cycles of manual revision using COOT (52) and crystallographic refinement with refmac5 (53). TLS refinement was used in the final stages, and omit maps were calculated to evaluate the accuracy of the model. Two copies of CTLA-4:ipilimumab complex were found in the crystallographic asymmetric unit (H/L, heavy chain/light chain; h/l, heavy chain/light chain; and C/c, two copies of CTLA-4). Electron density for both CTLA-4 molecules was continuous except for residues 63–64 of chain C and 64–65 of chain c. Residues 135–139 of chain H and 134–138 of chain h were also not modeled due to weak electron density. A few side chains near these regions were also truncated due to the lack of interpretable electron density. Residues populated in the favored, generously allowed, and disallowed regions in the Ramachandran plot account for 94.9%, 4.5%, and 0.6% of the total residues, respectively. The final model has been refined to an Rwork and Rfree of 20.3% and 26.7%, respectively, at a resolution of 3.0 Å (Table 1).

Structural Analysis.

Accessible surface area, superpositions, and atomic contacts were calculated with the areaimol (54) (www.ccp4.ac.uk/html/areaimol.html), lsqkab (55), and contact (56) programs in CCP4 Suite, respectively. Ribbon diagrams were generated with Pymol (57). The coordinates of the ipilimumab:CTLA-4 complex have been deposited in the PDB under ID code 5TRU.

SEC Analysis.

The interactions of wild-type dimeric CTLA-4 and dimeric mutant CTLA-4 with Fab and hB7-2 IgV were examined by SEC on Superdex 200 in buffer composed of 20 mM Hepes, pH 7.0, at 4 °C. Dimeric wild-type and mutant CTLA-4, Fab, and hB7-2 were individually chromatographed. CTLA-4 was incubated with different molar ratios of Fab or hB7-2 IgV and chromatographed.

Native PAGE Analysis.

Formation of the CTLA-4:Fab complex was demonstrated with 4–20% native gradient polyacrylamide gels (Bio-Rad). Dimeric CTLA-4 and Fab were mixed in different molar ratios and electrophoresed at 70 V for 7 h or 40 V for 12 h, at 4 °C, with buffer composed of 25 mM Tris·HCl and 250 mM glycine without SDS, pH 8.0. Gels were stained for 5 h with 0.25% Brilliant Blue R-250 (Fisher Biotech) in 50% MeOH and 10% HOAc and destained for 12 h in 30% MeOH and 10% HOAc. Formation of the dimeric CTLA-4:monomeric hB7-2 complexes were demonstrated by the same method.

Thermal Stability of CTLA-4 Mutants.

The fluorescence-monitored thermal denaturation of the dimeric wild-type and dimeric mutant CTLA-4 constructs was performed using an IQ5 thermalcycler (Bio-Rad). Briefly, 20 μL of protein at a concentration of 10 μM was mixed with 1 μL of Sypro-orange solution and pipetted into separate wells of a 96-well PCR plate. The temperature was increased from 20 to 99 °C, in 1 °C increments, with a dwell time of 6 s. The negative first derivative of the change in fluorescence (–dRFU/dT) for each protein was plotted against temperature, and the melting temperature was defined as the minimum value in the curve. Each measurement was made in duplicate. The errors of the melting temperatures were within 1 °C in all groups.

Site-Directed Mutagenesis to Generate an hCTLA-4 Mutant Library.

The coding sequence for the full-length ecto-domain of human CTLA-4 (Ala37-Phe162) was cloned into a ligation-independent cloning vector that adds the leader sequence from erythropoietin (EPO) and the transmembrane domain from mouse PD-L1 followed by mCherry. The resulting construct (EPO–hCTLA-4–mCherry) was transiently transfected into HEK293 suspension cells, and 2 d posttransfection, expression and correct membrane localization were verified by fluorescence microscopy. Site-specific mutagenesis was performed using high-fidelity KOD polymerase, 2 mM dNTPs, and 4 mM MgCl2. Positions selected for mutagenesis were based on the crystal structure of human CTLA-4 (PDB ID code 3OSK), which was used in calculating the surface accessibility with the online algorithm GetArea (58). All residues with a surface accessibility ratio >40% were selected for mutagenesis along with all charged residues that fell below that cutoff. Mutagenesis was attempted such that each chosen residue was mutated to an alanine and glutamic acid residue; acidic residues were mutated to arginine. The overall mutagenesis success rate was 80%, yielding 128 hCTLA-4 mutants for downstream assays. The sequence validated mutants were expression tested by transient transfection of 1 mL of suspension HEK293 cells. Only those mutants exhibiting comparable expression to the parental EPO–hCTLA-4–mCherry construct (by FACS) and correct plasma membrane localization (by fluorescence microscopy) were subsequently used in the FACS binding studies (118 total passed).

Epitope Mapping by High-Throughput FACS Analysis.

Wild-type and mutant hCTLA-4 constructs were transiently transfected into HEK293 suspension cells by plating 1 mL of cells at 1 × 106 cells per mL in 24-well plates and using 0.5 μg plasmid DNA and 2 μg linear PEI. Two days posttransfection, cells were counted and diluted to 1 × 106 cells per mL with 1x PBS with 2% BSA. We challenged 0.15 × 106 cells with 0.3 μg of ipilimumab (∼10 nM) for 1 h at room temperature while shaking in 96-well plates at 900 rpm on a MTS 2/4 microplate shaker (IKA). Cells were subsequently pelleted by centrifugation at 500 × g and washed three times with 1× PBS containing 2% BSA. Goat anti-human Alexa 488 secondary antibody (0.5 μg) was added to the cells, and incubated at 4 °C for 45 min. After washing three times, antibody binding was assessed by FACS analysis on a BD Accuri cytometer connected to an Intellicyt Hypercyte auto sampler. Flow data were gated for mCherry-positive events (hCTLA-4 expression) and then subgated for ipilimumab binding (Alexa 488 channel). The experiment was performed in triplicate, and the data from each experiment were normalized to wild type.

Binding of hB7-1, hB7-2, and hICOS-L to wild-type and mutant hCTLA-4 constructs was performed using a FACS cell–cell conjugation assay. HEK293 suspension cells were transiently transfected with “challenger” constructs for hB7-1, hB7-2, and hICOS-L expressed as C-terminal GFP fusions in parallel with hCTLA-4 mutant transfections. Two days posttransfection, each challenger cell population was diluted to 1 × 106 cells per mL with 1x PBS containing 2% BSA and mixed with a 1:1 ratio of HEK293 cells transiently transfected with each of the CTLA-4 mutant constructs. The cell–cell mixture was incubated at room temperature for 2 h while shaking in 96-well plates at 900 rpm on a MTS 2/4 microplate shaker (IKA) and subsequently analyzed by FACS as described above. FACS data were gated to determine the number of double-positive events [challenger-expressing cells (GFP) bound to hCTLA-4–expressing cells (mCherry)]. Percent bound was calculated by dividing the number of double-positive events by the total number of mCherry-positive events (total hCTLA-4 expression).

Bead-Based FACS Competition Assay.

We saturated 5 μL protein A beads (Bangs Labs) with hCTLA-4 hIgG1 (R&D Systems) at a concentration of 10 μg/μL beads. Bound beads were washed two times with 1× PBS containing 0.2% BSA by pelleting at 500 g and resuspended in 250 μL 1× PBS containing 0.2% BSA. We titrated 5μL of diluted beads loaded with CTLA-4 with increasing concentrations of hB7-1 mIgG2a protein (0–100 nM) (purified in house) in 100-μL reactions and allowed them to bind for 1 h at room temperature while shaking in 96-well plates at 900 rpm on a MTS 2/4 microplate shaker (IKA). Bead reactions were washed 3 times by centrifugation at 500 g using 1× PBS containing 0.2% BSA. Subsequently anti-mouse (H+L) Alexa 488 secondary antibody (0.005 μg/μL) was added and allowed to bind for 30 min. Reactions were washed an additional 3 times before FACS analysis. FACS data were gated for all beads and the GeoMean of FL-1 (488) for all beads was used to determine the amount of hB7-1 bound to the beads. For competition experiments, 5 µL of hCTLA-4–coated beads were added to reactions containing saturating hB7-1 mIgG2a (5 nM) and increasing concentrations (0–50 nM) of control or ipilimumab monoclonal antibodies. The reactions were washed and analyzed as described above, where loss of hB7-1 binding was indicative of competition.

Proteolytic Fragmentation for Epitope Peptide Analysis by MS.

MS epitope sequence analysis of ipilimumab was based on limited proteolysis (P1, P2). Purified ipilimumab was directly coupled onto surface-activated beads (Life Technologies) according to the supplier’s procedures. Intact human CTLA4-Fc and CD28-Fc antigens (R&D Systems) were used to generate peptides under both native and reduced conditions. Reduction of the antigen was performed by incubating with 50 mM DTT in PBS and 4 M guanidine HCl for 1 h at 37oC. This was followed by alkylation with 100 mM iodoacetamide for 20 min in the dark at room temperature. Reduced and alkylated antigens were dialyzed against PBS overnight before enzymatic proteolysis. Peptides were generated by proteolytic digestion using the endoproteinases trypsin, Lys-C, Arg-C, Asp-N, or Glu-C with an enzyme to antigen ratio of 2% (wt/wt) at 37 °C between 2 h to overnight. The resulting peptides were mixed with antibody beads for 30 min and followed by three washes with PBS before acid dissociation. Antibody-bound peptides were directly analyzed and identified by either MALDI-TOF or nano-LC–MS/MS.

Acknowledgments

We thank the staff of the 24-ID-E beamline, Advanced Photon Source, Argonne National Laboratory for assistance with X-ray diffraction data collection. This work was supported by National Institutes of Health Grants HG008325, GM094662, and GM094665 (to S.C.A.); and the Albert Einstein Cancer Center (P30CA013330).

Footnotes

  • ↵1U.A.R., W.L., and S.C.G.-T. contributed equally to this work.

  • ↵2Present address: Biological Sciences Division, Poornaprajna Institute of Scientific Research, Bangalore 562110, India.

  • ↵3Present address: Janssen Pharmaceuticals Inc., Titusville, NJ 08560.

  • ↵4Present address: F-Star Biotechnology Ltd., Cambridge, CB22 3AT, United Kingdom.

  • ↵5To whom correspondence may be addressed. Email: steve.almo{at}einstein.yu.edu or alan.korman{at}bms.com.

  • Author contributions: U.A.R., W.L., S.C.G.-T., J.B.B., Q.Y., M.S., S.C.W., A.B., S.M., V.S.R., S.D., A.J.K., and S.C.A. designed research; U.A.R., W.L., S.C.G.-T., Q.Y., M.S., S.C.W., A.B., S.M., V.S.R., and S.D. performed research; U.A.R., W.L., S.C.G.-T., J.B.B., Q.Y., M.S., S.C.W., A.B., S.M., V.S.R., and S.D. analyzed data; and U.A.R., W.L., S.C.G.-T., M.S., A.J.K., and S.C.A. wrote the paper.

  • Conflict of interest statement: S.C.A., S.C.G.-T., U.A.R., W.L., and Q.Y. declare no competing financial interests. A.B. is a former employee of Bristol–Myers & Squibb. A.J.K., M.S., S.C.W., S.M., V.S.R., and S.D. are employees and stockholders of Bristol–Myers & Squibb.

  • This article is a PNAS Direct Submission.

  • Data deposition: The crystallography, atomic coordinates, and structure factors have been deposited in the Protein Data Bank, www.rcsb.org/pdb/home/home.do (PDB ID code 5TRU).

  • This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1617941114/-/DCSupplemental.

Freely available online through the PNAS open access option.

References

  1. ↵
    1. Mellman I,
    2. Coukos G,
    3. Dranoff G
    (2011) Cancer immunotherapy comes of age. Nature 480:480–489.
    .
    OpenUrlCrossRefPubMed
  2. ↵
    1. Korman AJ,
    2. Peggs KS,
    3. Allison JP
    (2006) Checkpoint blockade in cancer immunotherapy. Adv Immunol 90:297–339.
    .
    OpenUrlCrossRefPubMed
  3. ↵
    1. McGranahan N, et al.
    (2016) Clonal neoantigens elicit T cell immunoreactivity and sensitivity to immune checkpoint blockade. Science 351:1463–1469.
    .
    OpenUrlAbstract/FREE Full Text
  4. ↵
    1. Couzin-Frankel J
    (2013) Breakthrough of the year 2013. Cancer immunotherapy. Science 342:1432–1433.
    .
    OpenUrlAbstract/FREE Full Text
  5. ↵
    1. Pardoll DM
    (2012) The blockade of immune checkpoints in cancer immunotherapy. Nat Rev Cancer 12:252–264.
    .
    OpenUrlCrossRefPubMed
  6. ↵
    1. Peggs KS,
    2. Quezada SA,
    3. Chambers CA,
    4. Korman AJ,
    5. Allison JP
    (2009) Blockade of CTLA-4 on both effector and regulatory T cell compartments contributes to the antitumor activity of anti-CTLA-4 antibodies. J Exp Med 206:1717–1725.
    .
    OpenUrlAbstract/FREE Full Text
  7. ↵
    1. Zou W,
    2. Wolchok JD,
    3. Chen L
    (2016) PD-L1 (B7-H1) and PD-1 pathway blockade for cancer therapy: Mechanisms, response biomarkers, and combinations. Sci Transl Med 8:328rv4.
    .
    OpenUrlFREE Full Text
  8. ↵
    1. Schwartzentruber DJ, et al.
    (2011) gp100 peptide vaccine and interleukin-2 in patients with advanced melanoma. N Engl J Med 364:2119–2127.
    .
    OpenUrlCrossRefPubMed
  9. ↵
    1. Metzler WJ, et al.
    (1997) Solution structure of human CTLA-4 and delineation of a CD80/CD86 binding site conserved in CD28. Nat Struct Biol 4:527–531.
    .
    OpenUrlCrossRefPubMed
  10. ↵
    1. Hodi FS, et al.
    (2010) Improved survival with ipilimumab in patients with metastatic melanoma. N Engl J Med 363:711–723.
    .
    OpenUrlCrossRefPubMed
  11. ↵
    1. Robert C, et al., KEYNOTE-006 investigators
    (2015) Pembrolizumab versus ipilimumab in advanced melanoma. N Engl J Med 372:2521–2532.
    .
    OpenUrlCrossRefPubMed
  12. ↵
    1. Wolchok JD, et al.
    (2013) Nivolumab plus ipilimumab in advanced melanoma. N Engl J Med 369:122–133.
    .
    OpenUrlCrossRefPubMed
  13. ↵
    1. Larkin J,
    2. Hodi FS,
    3. Wolchok JD
    (2015) Combined nivolumab and ipilimumab or monotherapy in untreated melanoma. N Engl J Med 373:1270–1271.
    .
    OpenUrlCrossRef
  14. ↵
    1. Postow MA, et al.
    (2015) Nivolumab and ipilimumab versus ipilimumab in untreated melanoma. N Engl J Med 372:2006–2017.
    .
    OpenUrlCrossRefPubMed
  15. ↵
    1. Antonia S, et al.
    (2016) Safety and antitumour activity of durvalumab plus tremelimumab in non-small cell lung cancer: A multicentre, phase 1b study. Lancet Oncol 17:299–308.
    .
    OpenUrlCrossRefPubMed
  16. ↵
    1. Qureshi OS, et al.
    (2011) Trans-endocytosis of CD80 and CD86: A molecular basis for the cell-extrinsic function of CTLA-4. Science 332:600–603.
    .
    OpenUrlAbstract/FREE Full Text
  17. ↵
    1. Corse E,
    2. Allison JP
    (2012) Cutting edge: CTLA-4 on effector T cells inhibits in trans. J Immunol 189:1123–1127.
    .
    OpenUrlAbstract/FREE Full Text
  18. ↵
    1. Wang CJ, et al.
    (2012) Cutting edge: Cell-extrinsic immune regulation by CTLA-4 expressed on conventional T cells. J Immunol 189:1118–1122.
    .
    OpenUrlAbstract/FREE Full Text
  19. ↵
    1. Quezada SA,
    2. Peggs KS,
    3. Curran MA,
    4. Allison JP
    (2006) CTLA4 blockade and GM-CSF combination immunotherapy alters the intratumor balance of effector and regulatory T cells. J Clin Invest 116:1935–1945.
    .
    OpenUrlCrossRefPubMed
  20. ↵
    1. Selby MJ, et al.
    (2013) Anti-CTLA-4 antibodies of IgG2a isotype enhance antitumor activity through reduction of intratumoral regulatory T cells. Cancer Immunol Res 1:32–42.
    .
    OpenUrlAbstract/FREE Full Text
  21. ↵
    1. Simpson TR, et al.
    (2013) Fc-dependent depletion of tumor-infiltrating regulatory T cells co-defines the efficacy of anti-CTLA-4 therapy against melanoma. J Exp Med 210:1695–1710.
    .
    OpenUrlAbstract/FREE Full Text
  22. ↵
    1. Ribas A, et al.
    (2013) Phase III randomized clinical trial comparing tremelimumab with standard-of-care chemotherapy in patients with advanced melanoma. J Clin Oncol 31:616–622.
    .
    OpenUrlAbstract/FREE Full Text
  23. ↵
    1. Schadendorf D, et al.
    (2015) Pooled analysis of long-term survival data from phase II and phase III trials of ipilimumab in unresectable or metastatic melanoma. J Clin Oncol 33:1889–1894.
    .
    OpenUrlAbstract/FREE Full Text
  24. ↵
    1. Eroglu Z, et al.
    (2015) Long term survival with cytotoxic T lymphocyte-associated antigen 4 blockade using tremelimumab. Eur J Cancer 51:2689–2697.
    .
    OpenUrl
  25. ↵
    1. Romano E, et al.
    (2015) Ipilimumab-dependent cell-mediated cytotoxicity of regulatory T cells ex vivo by nonclassical monocytes in melanoma patients. Proc Natl Acad Sci USA 112:6140–6145.
    .
    OpenUrlAbstract/FREE Full Text
  26. ↵
    1. Hodi FS, et al.
    (2008) Immunologic and clinical effects of antibody blockade of cytotoxic T lymphocyte-associated antigen 4 in previously vaccinated cancer patients. Proc Natl Acad Sci USA 105:3005–3010.
    .
    OpenUrlAbstract/FREE Full Text
  27. ↵
    1. Liakou CI, et al.
    (2008) CTLA-4 blockade increases IFNgamma-producing CD4+ICOShi cells to shift the ratio of effector to regulatory T cells in cancer patients. Proc Natl Acad Sci USA 105:14987–14992.
    .
    OpenUrlAbstract/FREE Full Text
  28. ↵
    1. Selby M, et al.
    (2013) Antitumor activity of concurrent blockade of immune checkpoint molecules CTLA-4 and PD-1 in preclinical models. J Clin Oncol 31:3061.
    .
    OpenUrlAbstract/FREE Full Text
  29. ↵
    1. Schwartz JC,
    2. Zhang X,
    3. Fedorov AA,
    4. Nathenson SG,
    5. Almo SC
    (2001) Structural basis for co-stimulation by the human CTLA-4/B7-2 complex. Nature 410:604–608.
    .
    OpenUrlCrossRefPubMed
  30. ↵
    1. Stamper CC, et al.
    (2001) Crystal structure of the B7-1/CTLA-4 complex that inhibits human immune responses. Nature 410:608–611.
    .
    OpenUrlCrossRefPubMed
  31. ↵
    1. Davies DR,
    2. Cohen GH
    (1996) Interactions of protein antigens with antibodies. Proc Natl Acad Sci USA 93:7–12.
    .
    OpenUrlAbstract/FREE Full Text
  32. ↵
    1. Jeon H, et al.
    (2014) Structure and cancer immunotherapy of the B7 family member B7x. Cell Reports 9:1089–1098.
    .
    OpenUrl
  33. ↵
    1. Hamby SE,
    2. Hirst JD
    (2008) Prediction of glycosylation sites using random forests. BMC Bioinformatics 9:500.
    .
    OpenUrlCrossRefPubMed
  34. ↵
    1. Zhang X,
    2. Schwartz JC,
    3. Almo SC,
    4. Nathenson SG
    (2002) Expression, refolding, purification, molecular characterization, crystallization, and preliminary X-ray analysis of the receptor binding domain of human B7-2. Protein Expr Purif 25:105–113.
    .
    OpenUrlCrossRefPubMed
  35. ↵
    1. Yu C, et al.
    (2011) Rigid-body ligand recognition drives cytotoxic T-lymphocyte antigen 4 (CTLA-4) receptor triggering. J Biol Chem 286:6685–6696.
    .
    OpenUrlAbstract/FREE Full Text
  36. ↵
    1. Evans EJ, et al.
    (2005) Crystal structure of a soluble CD28-Fab complex. Nat Immunol 6:271–279.
    .
    OpenUrlCrossRefPubMed
  37. ↵
    1. Schönfeld D, et al.
    (2009) An engineered lipocalin specific for CTLA-4 reveals a combining site with structural and conformational features similar to antibodies. Proc Natl Acad Sci USA 106:8198–8203.
    .
    OpenUrlAbstract/FREE Full Text
  38. ↵
    1. Ostrov DA,
    2. Shi W,
    3. Schwartz JC,
    4. Almo SC,
    5. Nathenson SG
    (2000) Structure of murine CTLA-4 and its role in modulating T cell responsiveness. Science 290:816–819.
    .
    OpenUrlAbstract/FREE Full Text
  39. ↵
    1. Chothia C,
    2. Novotný J,
    3. Bruccoleri R,
    4. Karplus M
    (1985) Domain association in immunoglobulin molecules. The packing of variable domains. J Mol Biol 186:651–663.
    .
    OpenUrlCrossRefPubMed
  40. ↵
    1. Almo SC,
    2. Guha C
    (2014) Considerations for combined immune checkpoint modulation and radiation treatment. Radiat Res 182:230–238.
    .
    OpenUrl
  41. ↵
    1. Liu X, et al.
    (2015) Affinity-tuned ErbB2 or EGFR chimeric antigen receptor T cells exhibit an increased therapeutic index against tumors in mice. Cancer Res 75:3596–3607.
    .
    OpenUrlAbstract/FREE Full Text
  42. ↵
    1. Caruso HG, et al.
    (2015) Tuning sensitivity of CAR to EGFR density limits recognition of normal tissue while maintaining potent antitumor activity. Cancer Res 75:3505–3518.
    .
    OpenUrlAbstract/FREE Full Text
  43. ↵
    1. Talavera A, et al.
    (2009) Nimotuzumab, an antitumor antibody that targets the epidermal growth factor receptor, blocks ligand binding while permitting the active receptor conformation. Cancer Res 69:5851–5859.
    .
    OpenUrlAbstract/FREE Full Text
  44. ↵
    1. Heitner T, et al.
    (2001) Selection of cell binding and internalizing epidermal growth factor receptor antibodies from a phage display library. J Immunol Methods 248:17–30.
    .
    OpenUrlCrossRefPubMed
  45. ↵
    1. Carter P, et al.
    (1992) Humanization of an anti-p185HER2 antibody for human cancer therapy. Proc Natl Acad Sci USA 89:4285–4289.
    .
    OpenUrlAbstract/FREE Full Text
  46. ↵
    1. Bellone M, et al.
    (2013) The acidity of the tumor microenvironment is a mechanism of immune escape that can be overcome by proton pump inhibitors. OncoImmunology 2:e22058.
    .
    OpenUrlPubMed
  47. ↵
    1. Estrella V, et al.
    (2013) Acidity generated by the tumor microenvironment drives local invasion. Cancer Res 73:1524–1535.
    .
    OpenUrlAbstract/FREE Full Text
  48. ↵
    1. Kato Y, et al.
    (2013) Acidic extracellular microenvironment and cancer. Cancer Cell Int 13:89.
    .
    OpenUrlCrossRefPubMed
  49. ↵
    1. Zhang X,
    2. Schwartz JC,
    3. Nathenson SG,
    4. Almo SC
    (2001) Crystallization and preliminary X-ray analysis of the complex between human CTLA-4 and B7-2. Acta Crystallogr D Biol Crystallogr 57:898–899.
    .
    OpenUrlCrossRefPubMed
  50. ↵
    1. Otwinowski Z,
    2. Minor W
    (1997) Processing of X-ray diffraction data collected in oscillation mode. Methods Enzymol 276:307–326.
    .
    OpenUrlCrossRefPubMed
  51. ↵
    1. Vagin A,
    2. Teplyakov A
    (2010) Molecular replacement with MOLREP. Acta Crystallogr D Biol Crystallogr 66:22–25.
    .
    OpenUrlCrossRefPubMed
  52. ↵
    1. Emsley P,
    2. Cowtan K
    (2004) Coot: Model-building tools for molecular graphics. Acta Crystallogr D Biol Crystallogr 60:2126–2132.
    .
    OpenUrlCrossRefPubMed
  53. ↵
    1. Murshudov GN,
    2. Vagin AA,
    3. Dodson EJ
    (1997) Refinement of macromolecular structures by the maximum-likelihood method. Acta Crystallogr D Biol Crystallogr 53:240–255.
    .
    OpenUrlCrossRefPubMed
  54. ↵
    1. Lee B,
    2. Richards FM
    (1971) The interpretation of protein structures: Estimation of static accessibility. J Mol Biol 55:379–400.
    .
    OpenUrlCrossRefPubMed
  55. ↵
    1. Kabsch W
    (1976) A solution for the best rotation to relate two sets of vectors. Acta Cryst A32:922–923.
    .
    OpenUrlCrossRef
  56. ↵
    1. Winn MD,
    2. et al.
    (2011) Overview of the CCP4 suite and current developments. Acta Cryst D67:235–242.
    .
    OpenUrlPubMed
  57. ↵
    1. DeLano WL
    (2005) The case for open-source software in drug discovery. Drug Discov Today 10:213–217.
    .
    OpenUrlCrossRefPubMed
  58. ↵
    1. Fraczkiewicz R,
    2. Braun WJ
    (1998) Exact and efficient analytical calculation of the accessible surface areas and their gradients for macromolecules. Comp Chem 19:319–333.
    .
    OpenUrl
View Abstract
Back to top
Article Alerts
Email Article
Citation Tools
Request Permissions
Structure of checkpoint inhibitor Ipilimumab
Udupi A. Ramagopal, Weifeng Liu, Sarah C. Garrett-Thomson, Jeffrey B. Bonanno, Qingrong Yan, Mohan Srinivasan, Susan C. Wong, Alasdair Bell, Shilpa Mankikar, Vangipuram S. Rangan, Shrikant Deshpande, Alan J. Korman, Steven C. Almo
Proceedings of the National Academy of Sciences May 2017, 114 (21) E4223-E4232; DOI: 10.1073/pnas.1617941114
del.icio.us logo Digg logo Reddit logo Twitter logo CiteULike logo Facebook logo Google logo Mendeley logo
Proceedings of the National Academy of Sciences: 116 (47)
Current Issue

Submit

Article Classifications

Jump to section

You May Also be Interested in

Comparisons of archival and modern data suggest the Thwaites Eastern Ice Shelf thinned by 10-33% between 1978 and 2009. Image courtesy of James Yungel/NASA ICE.
Radar data on Antarctic ice sheet
Comparisons of archival and modern data suggest the Thwaites Eastern Ice Shelf thinned by 10–33% between 1978 and 2009.
Image courtesy of James Yungel/NASA ICE.
Across most ancient and modern mammalian orders, male specimens outnumber female specimens in many museum collections. Image courtesy of ddouk/pixabay.
Bias in mammalian specimen collections
Across most ancient and modern mammalian orders, male specimens outnumber female specimens in many museum collections.
Image courtesy of Pixabay/ddouk.
QnAs with NAS member and parasitologist Patricia Johnson. Image courtesy of Joy Ahn (University of California, Los Angeles, CA).
Featured QnAs
QnAs with NAS member and parasitologist Patricia Johnson
Image courtesy of Joy Ahn (University of California, Los Angeles, CA).

Similar Articles