Testing of candidate monoclonal antibody therapeutics in preclinical models is an essential step in drug development. Identification of antibody therapeutic candidates that bind their human targets and cross-react to mouse orthologs is often challenging, especially for targets with low sequence homology. In such cases, surrogate antibodies that bind mouse orthologs must be used. The antibody 9D9, which binds mouse CTLA-4, is a commonly used surrogate for CTLA-4 checkpoint blockade studies in mouse cancer models. In this work, we reveal that 9D9 has significant biophysical dissimilarities to therapeutic CTLA-4 antibodies. The 9D9-mCTLA4 complex crystal structure was determined and shows that the surrogate antibody binds an epitope distinct from ipilimumab and tremelimumab. In addition, while ipilimumab has pH-independent binding to hCTLA-4, 9D9 loses binding to mCTLA-4 at physiologically relevant acidic pH ranges. We used phage and yeast display to engineer ipilimumab to bind mouse CTLA-4 with single-digit nM affinity from an initial state with no apparent binding. The engineered variants showed pH-independent and cross-reactive binding to both mouse and human CTLA-4. Crystal structures of a variant in complex with both mouse and human CTLA-4 confirmed that it targets an equivalent epitope as ipilimumab. These cross-reactive ipilimumab variants may facilitate improved translatability and future mechanism-of-action studies for anti-CTLA-4 targeting in murine models.

Design strategies that lead to a more focused in vivo delivery of functionalized nanoparticles (NPs) and their cargo can potentially maximize their therapeutic efficiency while reducing systemic effects, broadening their clinical applications. Here, we report the development of a noncovalent labeling approach where immunoglobulin G (IgG)-decorated NPs can be directed to a cancer cell using a simple, linear bispecific protein adaptor, termed MFE23-ZZ. MFE23-ZZ was created by fusing a single-chain fragment variable domain, termed MFE23, recognizing carcinoembryonic antigen (CEA) expressed on tumor cells, to a small protein ZZ module, which binds to the Fc fragment of IgG. As a proof of concept, monoclonal antibodies (mAbs) were generated against a NP coat protein, namely, gas vesicle protein A (GvpA) of Halobacterium salinarum gas vesicles (GVs). The surface of each GV was therapeutically derivatized with the photoreactive agent chlorin e6 (Ce6GVs) and anti-GvpA mAbs were subsequently bound to GvpA on the surface of each Ce6GV. The bispecific ligand MFE23-ZZ was then bound to mAb-decorated Ce6GVs via their Fc domain, resulting in a noncovalent tripartite complex, namely, MFE23.ZZ-2B10-Ce6GV. This complex enhanced the intracellular uptake of Ce6GVs into human CEA-expressing murine MC38 colon carcinoma cells (MC38.CEA) relative to the CEA-negative parental cell line MC38 in vitro, making them more sensitive to light-induced cell killing. These results suggest that the surface of NP can be rapidly and noncovalently functionalized to target tumor-associated antigen-expressing tumor cells using simple bispecific linkers and any IgG-labeled cargo. This noncovalent approach is readily applicable to other types of functionalized NPs.

The utility of antibody therapeutics is hampered by potential cross-reactivity with healthy tissue. Over the past decade, significant advances have been made in the design of activatable antibodies which increase, or create altogether, the therapeutic window of a parent antibody. Of these, antibody prodrugs (pro-antibodies) are masked antibodies that have advanced the most for therapeutic use. They are designed to reveal the active, parent antibody only when encountering proteases upregulated in the microenvironment of the targeted disease tissue, thereby minimizing off-target activity. However, current pro-antibody designs are relegated to fusion proteins that append masking groups restricted to the use of only canonical amino acids, offering excellent control of the site of introduction, but with no authority over where the masking group is installed other than the N-terminus of the antibody. Here, we present a palladium-based bioconjugation approach for the site-specific introduction of a masked tyrosine mimic in the complementary determining region of the FDA approved antibody therapeutic Ipilimumab used as a model system. The approach enables the introduction of a protease cleavable group tethered to non-canonical polymers (poly-ethylene glycol, PEG) resulting in 47-fold weaker binding to cells expressing CTLA-4, the target antigen of Ipilimumab. Upon exposure to tumor-associated proteases, the masking group is cleaved unveiling a tyrosine-mimic (dubbed a hydroxyphenyl cysteine, HPC) that restores (>90% restoration) binding affinity to its target antigen.

Caspases are a family of cysteine-dependent proteases with important cellular functions in inflammation and apoptosis, while also implicated in human diseases. Classical chemical tools to study caspase functions lack selectivity for specific caspase family members due to highly conserved active sites and catalytic machinery. To overcome this limitation, we targeted a non-catalytic cysteine residue (C264) unique to caspase-6 (C6), an enigmatic and understudied caspase isoform. Starting from disulfide ligands identified in a cysteine trapping screen, we used a structure-informed covalent ligand design to produce potent, irreversible inhibitors (3a) and chemoproteomic probes (13-t) of C6 that exhibit unprecedented selectivity over other caspase family members and high proteome selectivity. This approach and the new tools described will enable rigorous interrogation of the role of caspase-6 in developmental biology and in inflammatory and neurodegenerative diseases.

The clinical successes of immune checkpoint blockade have invigorated efforts to activate T cell-mediated responses against cancer. Targeting members of the PVR family, consisting of inhibitory receptors TIGIT, CD96, and CD112R, has been an active area of clinical investigation. In this study, the binding interactions and molecular assemblies of the PVR family receptors and ligands have been assessed in vitro. Furthermore, the anti-TIGIT monoclonal antibody BMS-986207 crystal structure in complex with TIGIT was determined and shows that the antibody binds an epitope that is commonly targeted by the CD155 ligand as well as other clinical anti-TIGIT antibodies. In contrast to previously proposed models, where TIGIT outcompetes costimulatory receptor CD226 for binding to CD155 due to much higher affinity (nanomolar range), our data rather suggest that PVR family members all engage in interactions with relatively weak affinity (micromolar range), including TIGIT and CD155 interactions. Thus, TIGIT and other PVR inhibitory receptors likely elicit immune suppression via increased surface expression rather than inherent differences in affinity. This work provides an improved foundational understanding of the PVR family network and mechanistic insight into therapeutic antibody intervention.

PVR (poliovirus receptor) functions as a ligand that signals through TIGIT and CD96 to induce suppression of T-cell and NK-cell responses. Alternatively, PVR binds to CD226, resulting in a co-stimulatory signal. To date, TIGIT antibody antagonists have been developed to restore immune functions and allow PVR to signal though CD226 in the context of cancer immunotherapy. Due to PVR receptor heterogeneity, agonizing either of these pathways with a recombinant form of the PVR extracellular domain represents a therapeutic strategy for either immunosuppression or activation. Here, we developed a minimal murine PVR-Fc fusion construct, consisting of only the IgV domain of PVR (vdPVR-Fc), and assessed its ability to dampen inflammatory responses in a murine model of psoriasis. vdPVR-Fc and PVR-Fc containing the full-length extracellular domain bound to TIGIT, CD96 and CD226 with similar low nanomolar affinities as defined by surface plasmon resonance. vdPVR-Fc was also able to suppress the in-vitro proliferation of murine CD4 + and CD8 + T-cells in mixed splenocyte cultures. Importantly, vdPVR-Fc delayed the onset, and reduced inflammatory responses (scaling and thickness) in a murine model of psoriasis. Collectively, our results suggest that the minimal IgV domain of PVR is sufficient to dampen immune responses in-vitro and attenuate symptoms of psoriasis in-vivo.

Characterization of antibody binding epitopes is an important factor in therapeutic drug discovery, as the binding site determines and drives antibody pharmacology and pharmacokinetics. Here, we present a novel application of carbene chemical footprinting with mass spectrometry for identification of antibody binding epitopes at the single-residue level. Two different photoactivated diazirine reagents provide complementary labeling information allowing structural refinement of the antibody binding interface. We applied this technique to map the epitopes of multiple MICA and CTLA-4 antibodies and validated the findings with X-ray crystallography and yeast surface display epitope mapping. The characterized epitopes were used to understand biolayer interferometry-derived competitive binding results at the structural level. We show that carbene footprinting provides fast and high-resolution epitope information critical in the antibody selection process and enables mechanistic understanding of function to accelerate the drug discovery process.

Although therapeutically efficacious, ipilimumab can exhibit dose-limiting toxicity that prevents maximal efficacious clinical outcomes and can lead to discontinuation of treatment. We hypothesized that an acidic pH-selective ipilimumab (pH Ipi), which preferentially and reversibly targets the acidic tumor microenvironment over the neutral periphery, may have a more favorable therapeutic index. While ipilimumab has pH-independent CTLA-4 affinity, pH Ipi variants have been engineered to have up to 50-fold enhanced affinity to CTLA-4 at pH 6.0 compared to pH 7.4. In hCTLA-4 knock-in mice, these variants have maintained anti-tumor activity and reduced peripheral activation, a surrogate marker for toxicity. pH-sensitive therapeutic antibodies may be a differentiating paradigm and a novel modality for enhanced tumor targeting and improved safety profiles.

The molecular interactions of mouse CD96 to CD155 ligand and to two surrogate antibodies have been investigated. Biophysical and structural studies demonstrate that CD96 forms a homodimer but assembles as 1:1 heterodimeric complexes with CD155 or with one of the surrogate antibodies, which compete for the same binding interface. In comparison, the other surrogate antibody binds across the mouse CD96 dimer and recognizes a quaternary epitope spanning both protomers to block exposure of the ligand-binding site. This study reveals different blocking mechanisms and modalities of these two antibodies and may provide insight into the functional effects of antibodies against CD96.

It is increasingly clear that the nectin family and its immunoreceptors shape the immune response to cancer through several pathways. Yet, even as antibodies against TIGIT, CD96, and CD112R advance into clinical development, biological and therapeutic questions remain unanswered. Here, we review recent progress, prospects, and challenges to understanding and tapping this family in cancer immunotherapy.

In order to design a conservation strategy for birds inhabiting the Great Western Woodlands (GWW) in southwestern Australia, we adopted a new approach for classifying birds into functional groups (BHFGs) based on an analysis of vegetation structure-related habitat resources (VHR). On the basis of hierarchical cluster analysis of the VHR variables we selected here, we selected a model that incorporated nine-BHFGs. This yielded a classification system that, due to our different input sources, is distinct from an existing foraging guild-based system. In conjunction with GIS technique, this new classification system has potential for effectively predicting and mapping landscape-scale habitat distribution.

Many viral surface glycoproteins and cell surface receptors are homo-oligomers, and thus can potentially be targeted by geometrically matched homo-oligomers that engage all subunits simultaneously to attain high avidity and/or lock subunits together. The adaptive immune system cannot generally employ this strategy since the individual antibody binding sites are not arranged with appropriate geometry to simultaneously engage multiple sites in a single target homo-oligomer. We describe a general strategy for the computational design of homo-oligomeric protein assemblies with binding functionality precisely matched to homo-oligomeric target sites. In the first step, a small protein is designed that binds a single site on the target. In the second step, the designed protein is assembled into a homo-oligomer such that the designed binding sites are aligned with the target sites. We use this approach to design high-avidity trimeric proteins that bind influenza A hemagglutinin (HA) at its conserved receptor binding site. The designed trimers can both capture and detect HA in a paper-based diagnostic format, neutralizes influenza in cell culture, and completely protects mice when given as a single dose 24 h before or after challenge with influenza.

Cysteine can be specifically functionalized by a myriad of acid-base conjugation strategies for applications ranging from probing protein function to antibody-drug conjugates and proteomics. In contrast, selective ligation to the other sulfur-containing amino acid, methionine, has been precluded by its intrinsically weaker nucleophilicity. Here, we report a strategy for chemoselective methionine bioconjugation through redox reactivity, using oxaziridine-based reagents to achieve highly selective, rapid, and robust methionine labeling under a range of biocompatible reaction conditions. We highlight the broad utility of this conjugation method to enable precise addition of payloads to proteins, synthesis of antibody-drug conjugates, and identification of hyperreactive methionine residues in whole proteomes.

Human and avian influenza viruses recognize different sialic acid-containing receptors, referred to as human-type (NeuAcα2-6Gal) and avian-type (NeuAcα2-3Gal), respectively. This presents a species barrier for aerosol droplet transmission of avian viruses in humans and ferrets. Recent reports have suggested that current human H3N2 viruses no longer have strict specificity toward human-type receptors. Using an influenza receptor glycan microarray with extended airway glycans, we find that H3N2 viruses have in fact maintained human-type specificity, but they have evolved preference for a subset of receptors comprising branched glycans with extended poly-N-acetyl-lactosamine (poly-LacNAc) chains, a specificity shared with the 2009 pandemic H1N1 (Cal/04) hemagglutinin. Lipid-linked versions of extended sialoside receptors can restore susceptibility of sialidase-treated MDCK cells to infection by both recent (A/Victoria/361/11) and historical (A/Hong Kong/8/1968) H3N2 viruses. Remarkably, these human-type receptors with elongated branches have the potential to increase avidity by simultaneously binding to two subunits of a single hemagglutinin trimer.

Broadly neutralizing antibodies targeting a highly conserved region in the hemagglutinin (HA) stem protect against influenza infection. Here, we investigate the protective efficacy of a protein (HB36.6) computationally designed to bind with high affinity to the same region in the HA stem. We show that intranasal delivery of HB36.6 affords protection in mice lethally challenged with diverse strains of influenza independent of Fc-mediated effector functions or a host antiviral immune response. This designed protein prevents infection when given as a single dose of 6.0 mg/kg up to 48 hours before viral challenge and significantly reduces disease when administered as a daily therapeutic after challenge. A single dose of 10.0 mg/kg HB36.6 administered 1-day post-challenge resulted in substantially better protection than 10 doses of oseltamivir administered twice daily for 5 days. Thus, binding of HB36.6 to the influenza HA stem region alone, independent of a host response, is sufficient to reduce viral infection and replication in vivo. These studies demonstrate the potential of computationally designed binding proteins as a new class of antivirals for influenza.

Arylfluorosulfates have appeared only rarely in the literature and have not been explored as probes for covalent conjugation to proteins, possibly because they were assumed to possess high reactivity, as with other sulfur(VI) halides. However, we find that arylfluorosulfates become reactive only under certain circumstances, e.g., when fluoride displacement by a nucleophile is facilitated. Herein, we explore the reactivity of structurally simple arylfluorosulfates toward the proteome of human cells. We demonstrate that the protein reactivity of arylfluorosulfates is lower than that of the corresponding aryl sulfonyl fluorides, which are better characterized with regard to proteome reactivity. We discovered that simple hydrophobic arylfluorosulfates selectively react with a few members of the intracellular lipid binding protein (iLBP) family. A central function of iLBPs is to deliver small-molecule ligands to nuclear hormone receptors. Arylfluorosulfate probe 1 reacts with a conserved tyrosine residue in the ligand-binding site of a subset of iLBPs. Arylfluorosulfate probes 3 and 4, featuring a biphenyl core, very selectively and efficiently modify cellular retinoic acid binding protein 2 (CRABP2), both in vitro and in living cells. The X-ray crystal structure of the CRABP2-4 conjugate, when considered together with binding site mutagenesis experiments, provides insight into how CRABP2 might activate arylfluorosulfates toward site-specific reaction. Treatment of breast cancer cells with probe 4 attenuates nuclear hormone receptor activity mediated by retinoic acid, an endogenous client lipid of CRABP2. Our findings demonstrate that arylfluorosulfates can selectively target single iLBPs, making them useful for understanding iLBP function.

Caspases are a family of proteases found in all metazoans, including a dozen in humans, that drive the terminal stages of apoptosis as well as other cellular remodeling and inflammatory events. Caspases are named because they are cysteine class enzymes shown to cleave after aspartate residues. In the past decade, we and others have developed unbiased proteomic methods that collectively identified ~2000 native proteins cleaved during apoptosis after the signature aspartate residues. Here, we explore non-aspartate cleavage events and identify 100s of substrates cleaved after glutamate in both human and murine apoptotic samples. The extended consensus sequence patterns are virtually identical for the aspartate and glutamate cleavage sites suggesting they are cleaved by the same caspases. Detailed kinetic analyses of the dominant apoptotic executioner caspases-3 and -7 show that synthetic substrates containing DEVD↓ are cleaved only twofold faster than DEVE↓, which is well within the 500-fold range of rates that natural proteins are cut. X-ray crystallography studies confirm that the two acidic substrates bind in virtually the same way to either caspases-3 or -7 with minimal adjustments to accommodate the larger glutamate. Lastly, during apoptosis we found 121 proteins cleaved after serine residues that have been previously annotated to be phosphorylation sites. We found that caspase-3, but not caspase-7, can cleave peptides containing DEVpS↓ at only threefold slower rate than DEVD↓, but does not cleave the unphosphorylated serine peptide. There are only a handful of previously reported examples of proteins cleaved after glutamate and none after phosphorserine. Our studies reveal a much greater promiscuity for cleaving after acidic residues and the name 'cacidase' could aptly reflect this broader specificity.