The Tmod cellular logic gate as a solution for tumor-selective immunotherapy.

Immune cells that are engineered with receptors to integrate signals from multiple antigens offer a promising route to achieve the elusive property of therapeutic selectivity in cancer patients. Several types of multi-signal integrators have been described, among them mechanisms that pair activating and inhibitory receptors which are termed NOT gates by analogy to logical operations performed by machines. Here we review one such NOT-gated signal integrator called the Tmod system which is being developed for patients with solid tumors. Coupled with rigorous selection for patients with defined lesions in their tumor genomes (loss of heterozygosity), the Tmod approach presents an unusual opportunity to create truly selective therapies for certain cancer patients. Several of these agents are advancing toward the clinic, supported by a large body of quantitative preclinical data.

Geometric parameters that affect the behavior of logic-gated CAR T cells.

Clinical applications of CAR-T cells are limited by the scarcity of tumor-specific targets and are often afflicted with the same on-target/off-tumor toxicities that plague other cancer treatments. A new promising strategy to enforce tumor selectivity is the use of logic-gated, two-receptor systems. One well-described application is termed Tmod™, which originally utilized a blocking inhibitory receptor directed towards HLA-I target antigens to create a protective NOT gate. Here we show that the function of Tmod blockers targeting non-HLA-I antigens is dependent on the height of the blocker antigen and is generally compatible with small, membrane-proximal targets. We compensate for this apparent limitation by incorporating modular hinge units to artificially extend or retract the ligand-binding domains relative to the effector cell surface, thereby modulating Tmod activator and blocker function. By accounting for structural differences between activator and blocker targets, we developed a set of simple geometric parameters for Tmod receptor design that enables targeting of blocker antigens beyond HLA-I, thereby broadening the applications of logic-gated cell therapies.

HLA-A∗02-gated safety switch for cancer therapy has exquisite specificity for its allelic target antigen.

Innovative cell-based therapies are important new weapons in the fight against difficult-to-treat cancers. One promising strategy involves cell therapies equipped with multiple receptors to integrate signals from more than one antigen. We developed a specific embodiment of this approach called Tmod, a two-receptor system that combines activating and inhibitory inputs to distinguish between tumor and normal cells. The selectivity of Tmod is enforced by the inhibitory receptor (blocker) that recognizes an antigen, such as an HLA allele, whose expression is absent from tumors because of loss of heterozygosity. Although unwanted cross-reactivity of the blocker likely reduces efficacy rather than safety, it is important to verify the blocker's specificity. We have tested an A∗02-directed blocker derived from the PA2.1 mouse antibody as a safety mechanism paired with a mesothelin-specific activating CAR in our Tmod construct. We solved the crystal structure of humanized PA2.1 Fab in complex with HLA-A∗02 to determine its binding epitope, which was used to bioinformatically select specific class I HLA alleles to test the blocker's functional specificity in vitro. We found that this A∗02-directed blocker is highly specific for its cognate antigen, with only one cross-reactive allele (A∗69) capable of triggering comparable function.

Chimeric Antigen Receptors Directed at Mutant KRAS Exhibit an Inverse Relationship Between Functional Potency and Neoantigen Selectivity.

Neoantigens are among the most intriguing potential immuno-oncology targets because, unlike many cancer targets that are expressed on normal tissues, they are by definition restricted to cancer cells. Medicines directed at common neoantigens such as mutant KRAS are especially interesting because they may offer the convenience and cost of an off-the-shelf therapy. However, all common KRAS mutations produce proteins that differ from the wild type at a single amino acid, creating challenges for molecular discrimination. We have undertaken an effort to optimize single-chain variable fragments (scFv) against peptide/major histocompatibility antigen complexes composed of HLA-A*11 and either G12V- or G12D-mutant KRAS peptides. These scFvs could in principle be used in chimeric antigen receptor (CAR) T-cell therapies for selected patients whose tumors bear either of these mutations. Here we show that optimization of such CARs involves a trade-off between potency and selectivity. We further show that targeting this family without high selectivity engenders risks of cross-reactivity against other members of the G-protein family to which KRAS belongs. Significance: We report an effort to generate high potency, selective CARs directed at mutant KRAS peptides. Although the heavily optimized CARs maintain high selectivity against wild-type KRAS, they lose selectivity against other KRAS-related peptides derived from human proteins. To our knowledge, this work is the first to examine the trade-off between potency and selectivity with regard to KRAS pMHC-directed CARs, illustrating the challenge to achieve both sufficient potency and high selectivity.

Mesothelin-specific CAR-T cell therapy that incorporates an HLA-gated safety mechanism selectively kills tumor cells.

BACKGROUND: Mesothelin (MSLN) is a classic tumor-associated antigen that is expressed in lung cancer and many other solid tumors. However, MSLN is also expressed in normal mesothelium which creates a significant risk of serious inflammation for MSLN-directed therapeutics. We have developed a dual-receptor (Tmod™) system that exploits the difference between tumor and normal tissue in a subset of patients with defined heterozygous gene loss (LOH) in their tumors. METHODS: T cells engineered with the MSLN CAR Tmod construct described here contain (1) a novel MSLN-activated CAR and (2) an HLA-A*02-gated inhibitory receptor (blocker). A*02 binding is intended to override T-cell cytotoxicity, even in the presence of MSLN. The Tmod system is designed to treat heterozygous HLA class I patients, selected for HLA LOH. When A*02 is absent from tumors selected for LOH, the MSLN Tmod cells are predicted to mediate potent killing of the MSLN(+)A*02(-) malignant cells. RESULTS: The sensitivity of the MSLN Tmod cells is comparable with a benchmark MSLN CAR-T that was active but toxic in the clinic. Unlike MSLN CAR-T cells, the Tmod system robustly protects surrogate "normal" cells even in mixed-cell populations in vitro and in a xenograft model. The MSLN CAR can also be paired with other HLA class I blockers, supporting extension of the approach to patients beyond A*02 heterozygotes. CONCLUSIONS: The Tmod mechanism exemplified by the MSLN CAR Tmod construct provides an alternative route to leverage solid-tumor antigens such as MSLN in safer, more effective ways than previously possible.

Engineered T cells directed at tumors with defined allelic loss.

We describe an approach to cancer therapy based on exploitation of common losses of genetic material in tumor cells (loss of heterozygosity) (Basilion et al., 1999; Beroukhim et al., 2010). This therapeutic concept addresses the fundamental problem of discrimination between tumor and normal cells and can be applied in principle to the large majority of tumors. It utilizes modular activator/blocker elements that integrate signals related to the presence and absence of ligands displayed on the cell surface (Fedorov et al., 2013). We show that the targeting system works robustly in vitro and in a mouse cancer model where absence of the HLA-A*02 allele releases a brake on engineered T cells activated by the CD19 surface antigen. This therapeutic approach potentially opens a route toward a large, new source of cancer targets.

Structure-function relationships of chimeric antigen receptors in acute T cell responses to antigen.

Chimeric antigen receptors (CARs) and their parent signaling molecule, the T cell receptor (TCR), are fascinating proteins of increasing relevance to disease therapy. Here we use a collection of 1221 pMHC-directed CAR constructs representing 10 pMHC targets to study aspects of CAR structure-activity relationships (SAR), with particular focus on the extracellular and transmembrane structural components. These experiments that involve pMHC targets whose number/cell can be manipulated by peptide dosing in vitro enable systematic analysis of the SAR of CARs in carefully controlled experimental situations (Harris and Kranz, 2016). We find that CARs tolerate a wide range of structural variation, with the ligand-binding domains (LBDs) dominating the SAR of CAR antigen sensitivity. Notwithstanding the critical role of the LBD, CAR antigen-binding on the cell surface, measured by pMHC tetramer staining, is not an effective predictor of functional sensitivity. These results have important implications for the design and testing of CARs aimed toward the clinic.

Differences in binding and effector functions between classes of TNF antagonists.

There are currently two Food and Drug Administration-approved classes of biologic agents that target tumor necrosis factor-alpha (TNF-alpha): anti-TNF monoclonal antibodies (mAbs) (adalimumab and infliximab), and soluble TNF receptors (etanercept). This study examined the ability of the TNF antagonists to: (1) bind various polymorphic variants of cell surface-expressed Fc receptors (FcgammaRs) and the complement component C1q, and (2) mediate Ab-dependent cellular cytotoxicity (ADCC) and complement-mediated cytotoxicity (CDC) killing of cells expressing membrane-bound TNF (mTNF) in vitro. Both mAbs and the soluble TNF receptor demonstrated low-level binding to the activating receptors FcgammaRI, FcgammaRIIa, and FcgammaRIIIa, and the inhibitory receptor FcgammaRIIb, in the absence of exogenous TNF. However, upon addition of TNF, the mAbs, but not etanercept, showed significantly increased binding, in particular to the FcgammaRII and FcgammaRIII receptors. Infliximab and adalimumab induced ADCC much more potently than etanercept. In the presence of TNF, both mAbs bound C1q in in vitro assays, but etanercept did not bind C1q under any conditions. Infliximab and adalimumab also induced CDC in cells expressing mTNF more potently than etanercept. Differences in the ability to bind ligand and mediate cell death may account for the differences in efficacy and safety of TNF antagonists.

Crystal structure of the S.cerevisiae exocyst component Exo70p.

The exocyst is an evolutionarily conserved multiprotein complex required for the targeting and docking of post-Golgi vesicles to the plasma membrane. Through its interactions with a variety of proteins, including small GTPases, the exocyst is thought to integrate signals from the cell and signal that vesicles arriving at the plasma membrane are ready for fusion. Here we describe the three-dimensional crystal structure of one of the components of the exocyst, Exo70p, from Saccharomyces cerevisiae at 3.5A resolution. Exo70p binds the small GTPase Rho3p in a GTP-dependent manner with an equilibrium dissociation constant of approximately 70 microM. Exo70p is an extended rod approximately 155 angstroms in length composed principally of alpha helices, and is a novel fold. The structure provides a first view of the Exo70 protein family and provides a framework to study the molecular function of this exocyst component.

Crystal structure of a secreted insect ferritin reveals a symmetrical arrangement of heavy and light chains.

Ferritins are iron storage proteins made of 24 subunits forming a hollow spherical shell. Vertebrate ferritins contain varying ratios of heavy (H) and light (L) chains; however, known ferritin structures include only one type of chain and have octahedral symmetry. Here, we report the 1.9A structure of a secreted insect ferritin from Trichoplusia ni, which reveals equal numbers of H and L chains arranged with tetrahedral symmetry. The H/L-chain interface includes complementary features responsible for ordered assembly of the subunits. The H chain contains a ferroxidase active site resembling that of vertebrate H chains with an endogenous, bound iron atom. The L chain lacks the residues that form a putative iron core nucleation site in vertebrate L chains. Instead, a possible nucleation site is observed at the L chain 3-fold pore. The structure also reveals inter- and intrasubunit disulfide bonds, mostly in the extended N-terminal regions unique to insect ferritins. The symmetrical arrangement of H and L chains and the disulfide crosslinks reflect adaptations of insect ferritin to its role as a secreted protein.

Crystal structure of a polymeric immunoglobulin binding fragment of the human polymeric immunoglobulin receptor.

The polymeric immunoglobulin receptor (pIgR) is a type I transmembrane protein that delivers dimeric IgA (dIgA) and pentameric IgM to mucosal secretions. Here, we report the 1.9 A resolution X-ray crystal structure of the N-terminal domain of human pIgR, which binds dIgA in the absence of other pIgR domains with an equilibrium dissociation constant of 300 nM. The structure of pIgR domain 1 reveals a folding topology similar to immunoglobulin variable domains, but with differences in the counterparts of the complementarity determining regions (CDRs), including a helical turn in CDR1 and a CDR3 loop that points away from the other CDRs. The unusual CDR3 loop position prevents dimerization analogous to the pairing of antibody variable heavy and variable light domains. The pIgR domain 1 structure allows interpretation of previous mutagenesis results and structure-based comparisons between pIgR and other IgA receptors.

Steric accessibility of the HIV-1 gp41 N-trimer region.

During human immunodeficiency virus entry, gp41 undergoes a series of conformational changes that induce membrane fusion. Immediately prior to fusion, gp41 exists in a prehairpin intermediate in which the N- and C-peptide regions of gp41 are exposed. Rearrangement of this intermediate into a six-helix bundle composed of a trimeric coiled coil from the N-peptide region (N-trimer) surrounded by three peptides from the C-peptide region provides the driving force for membrane fusion, whereas prevention of six-helix bundle formation inhibits viral entry. Because of its central role in mediating viral entry, the N-trimer region of gp41 is a key vaccine target. Extensive efforts to discover potent and broadly neutralizing antibodies (Abs) against the N-trimer region have, thus far, been unsuccessful. In this study, we attached a potent C-peptide inhibitor that binds to the N-trimer region to cargo proteins of various sizes to examine the steric accessibility of the N-trimer during fusion. These inhibitors show a progressive loss of potency with increasing cargo size. Extension of the cargo/C-peptide linker partially restores inhibitory potency. These results demonstrate that the human immunodeficiency virus defends its critical hairpin-forming machinery by steric exclusion of large proteins and may explain the current dearth of neutralizing Abs against the N-trimer. In contrast, previous results suggest the C-peptide region is freely accessible during fusion, demonstrating that the N- and C-peptide regions are in structurally distinct environments. Based on these results, we also propose new strategies for the generation of neutralizing Abs that overcome this steric block.

Solution structure of choline binding protein A, the major adhesin of Streptococcus pneumoniae.

Streptococcus pneumoniae (pneumococcus) remains a significant health threat worldwide, especially to the young and old. While some of the biomolecules involved in pneumococcal pathogenesis are known and understood in mechanistic terms, little is known about the molecular details of bacterium/host interactions. We report here the solution structure of the 'repeated' adhesion domains (domains R1 and R2) of the principal pneumococcal adhesin, choline binding protein A (CbpA). Further, we provide insights into the mechanism by which CbpA binds its human receptor, polymeric immunoglobulin receptor (pIgR). The R domains, comprised of 12 imperfect copies of the leucine zipper heptad motif, adopt a unique 3-alpha-helix, raft-like structure. Each pair of alpha-helices is antiparallel and conserved residues in the loop between Helices 1 and 2 exhibit a novel 'tyrosine fork' structure that is involved in binding pIgR. This and other structural features that we show are conserved in most pneumococcal strains appear to generally play an important role in bacterial adhesion to pIgR. Interestingly, pneumococcus is the only bacterium known to adhere to and invade human cells by binding to pIgR.