Genome-scale functional genomics screening highlights genes impacting protein fucosylation in Chinese hamster ovary cells.

N-linked glycosylation is a common post-translational modification that has various effects on multiple types of proteins. The extent to which an N-linked glycoprotein is modified and the identity of glycans species involved is of great interest to the biopharmaceutical industry, since glycosylation can impact the efficacy and safety of therapeutic monoclonal antibodies (mAbs). mAbs lacking core fucose, for example, display enhanced clinical efficacy through increased antibody-dependent cellular cytotoxicity. We performed a genome-wide CRISPR knockout screen in Chinese hamster ovary (CHO) cells, the workhorse cell culture system for industrial production of mAbs, aimed at identifying novel regulators of protein fucosylation. Using a lectin binding assay, we identified 224 gene perturbations that significantly alter protein fucosylation, including well-known glycosylation genes. This functional genomics framework could readily be extended and applied to study the genetic pathways involved in regulation of other glycoforms. We hope this resource will provide useful guidance toward the development of next generation CHO cell lines and mAb therapeutics.

Novel method to specifically determine the structures of non-N297 glycans in IgGs.

Monoclonal antibodies comprise a major class of biologic therapeutics and are also extensively studied in immunology. Given the importance of glycans on antibodies, fluorescent labeling of enzymatically released glycans and their LC/MS analysis is routinely used for in-depth characterization of antibody glycosylation. In this technical note, we propose a method for facile characterization of glycans in the variable region of antibodies using sequential enzymatic digests with Endoglycosidase-S2 and RapidTM Peptide-N-Glycosidase-F followed by labeling with a fluorescent dye carrying an NHS-carbamate moiety. The results and proposed mechanism also suggest that the choice of glycosidases along with the labeling chemistry is critical for accurate glycan analysis for a desired application.

In silico high throughput mutagenesis and screening of signal peptides to mitigate N-terminal heterogeneity of recombinant monoclonal antibodies.

N-terminal heterogeneity resulting from non-uniform signal peptide (SP) cleavage can potentially affect biologics property attributes and result in extended product development timelines. Few studies are available on engineering SPs systematically to address miscleavage issues. Herein, we developed a novel high throughput computational pipeline capable of generating millions of SP mutant sequences that uses the SignalP 5.0 deep learning model to predict which of these mutants are likely to alleviate the N-terminal miscleavage in antibodies. We optimized the parameters to target mutating one or two amino acids at the C-terminus of 84 unique SPs, exhausting all theoretically possible combinations and resulting in a library of 296,077 unique wildtype and mutant signal peptides for in silico screening of each antibody. We applied this method to five antibodies against different targets, with various extent of miscleavage (2.3% to 100%) on their Lambda light chains. In each case, multiple SP mutants were generated, with miscleavage reduced to a non-detectable level and titers comparable with or better than that of the original SPs. Pairwise mutational analysis using an in silico library enriched with high-scoring mutants revealed patterns of amino acids at the C-terminus of SPs, providing insights beyond the "Heijne rule". To our knowledge, no similar approach that combines high throughput in silico mutagenesis and screening with SP cleavage prediction has been reported in the literature. This method can be applied to both the light chain and heavy chain of antibodies, regardless of their initial extent of miscleavage, provides optimized solutions for individual cases, and facilitates the development of antibody therapeutics.Abbreviations: Aa, amino acids; CHO, Chinese hamster ovary; CNN, convolutional neural network; CSscore, cleavage site score; CSV, comma-separated values; HC, heavy chain; HEK, human embryonic kidney; HPLC, high-performance liquid chromatography; IgG, immunoglobulin G; IGLV, immunoglobulin G Lambda variable; LC, light chain; LCMS, liquid chromatography-mass spectrometry; MS, mass spectrometry; PCR, polymerase chain reaction; PBS, phosphate-buffered saline; PEI, polyethylenimine; SP, signal peptide; SPase, signal peptidase; TCEP, tris(2-carboxyethyl) phosphine; TOF, time-of-flight.

An anti-PD-1-GITR-L bispecific agonist induces GITR clustering-mediated T cell activation for cancer immunotherapy.

Costimulatory receptors such as glucocorticoid-induced tumor necrosis factor receptor-related protein (GITR) play key roles in regulating the effector functions of T cells. In human clinical trials, however, GITR agonist antibodies have shown limited therapeutic effect, which may be due to suboptimal receptor clustering-mediated signaling. To overcome this potential limitation, a rational protein engineering approach is needed to optimize GITR agonist-based immunotherapies. Here we show a bispecific molecule consisting of an anti-PD-1 antibody fused with a multimeric GITR ligand (GITR-L) that induces PD-1-dependent and FcγR-independent GITR clustering, resulting in enhanced activation, proliferation and memory differentiation of primed antigen-specific GITR+PD-1+ T cells. The anti-PD-1-GITR-L bispecific is a PD-1-directed GITR-L construct that demonstrated dose-dependent, immunologically driven tumor growth inhibition in syngeneic, genetically engineered and xenograft humanized mouse tumor models, with a dose-dependent correlation between target saturation and Ki67 and TIGIT upregulation on memory T cells. Anti-PD-1-GITR-L thus represents a bispecific approach to directing GITR agonism for cancer immunotherapy.

Empowering therapeutic antibodies with IFN-α for cancer immunotherapy.

Type 1 IFNs stimulate secretion of IP-10 (CXCL10) which is a critical chemokine to recruit effector T cells to the tumor microenvironment and IP-10 knockout mice exhibit a phenotype with compromised effector T cell generation and trafficking. Type 1 IFNs also induce MHC class 1 upregulation on tumor cells which can enhance anti-tumor CD8 T cell effector response in the tumor microenvironment. Although type 1 IFNs show great promise in potentiating anti-tumor immune response, systemic delivery of type 1 IFNs is associated with toxicity thereby limiting clinical application. In this study, we fused tumor targeting antibodies with IFN-α and showed that the fusion proteins can be produced with high yields and purity. IFN fusions selectively induced IP-10 secretion from antigen positive tumor cells, which was critical in recruiting the effector T cells to the tumor microenvironment. Further, we found that treatment with the anti-PDL1-IFN- α fusion at concentrations as low as 1 pM exhibited potent activity in mediating OT1 CD8+ T cell killing against OVA expressing tumor cells, while control IFN fusion did not exhibit any activity at the same concentration. Furthermore, the IFN-α fusion antibody was well tolerated in vivo and demonstrated anti-tumor efficacy in an anti-PD-L1 resistant syngeneic mouse tumor model. One of the potential mechanisms for the enhanced CD8 T cell killing by anti-PD-L1 IFN fusion was up-regulation of MHC class I/tumor antigen complex. Our data supports the hypothesis of targeting type 1 IFN to the tumor microenvironment may enhance effector T cell functions for anti-tumor immune response.

A Bispecific Molecule Targeting CD40 and Tumor Antigen Mesothelin Enhances Tumor-Specific Immunity.

Agonistic CD40 monoclonal antibodies (mAb) have demonstrated some clinical activity, but with dose-limiting toxicity. To reduce systemic toxicity, we developed a bispecific molecule that was maximally active in the presence of a tumor antigen and had limited activity in the absence of the tumor antigen. LB-1 is a bispecific molecule containing single-chain Fv domains targeting mouse CD40 and the tumor antigen mesothelin. LB-1 exhibited enhanced activity upon binding to cell-surface mesothelin but was less potent in the absence of mesothelin binding. In a mouse model implanted with syngeneic 4T1 tumors expressing cell-surface mesothelin, LB-1 demonstrated comparable antitumor activity as an agonistic CD40 mAb but did not cause elevation of serum cytokines and liver enzymes, as was observed in anti-CD40-treated mice. The results from our study of LB-1 were used to develop a human cross-reactive bispecific molecule (ABBV-428) that targeted human CD40 and mesothelin. ABBV-428 demonstrated enhanced activation of antigen-presenting cells and T cells upon binding to cell-surface mesothelin, and inhibition of cultured or implanted PC3 tumor cell growth after immune activation. Although expression of cell-surface mesothelin is necessary, the bispecific molecules induced immune-mediated antitumor activity against both mesothelin+ and mesothelin- tumor cells. ABBV-428 represents a class of bispecific molecules with conditional activity dependent on the binding of a tumor-specific antigen, and such activity could potentially maximize antitumor potency while limiting systemic toxicity in clinical studies.

CD40/anti-CD40 antibody complexes which illustrate agonist and antagonist structural switches.

BACKGROUND: CD40 is a 48 kDa type I transmembrane protein that is constitutively expressed on hematopoietic cells such as dendritic cells, macrophages, and B cells. Engagement of CD40 by CD40L expressed on T cells results in the production of proinflammatory cytokines, induces T helper cell function, and promotes macrophage activation. The involvement of CD40 in chronic immune activation has resulted in CD40 being proposed as a therapeutic target for a range of chronic inflammatory diseases. CD40 antagonists are currently being explored for the treatment of autoimmune diseases and several anti-CD40 agonist mAbs have entered clinical development for oncological indications. RESULTS: To better understand the mode of action of anti-CD40 mAbs, we have determined the x-ray crystal structures of the ABBV-323 (anti-CD40 antagonist, ravagalimab) Fab alone, ABBV-323 Fab complexed to human CD40 and FAB516 (anti-CD40 agonist) complexed to human CD40. These three crystals structures 1) identify the conformational CD40 epitope for ABBV-323 recognition 2) illustrate conformational changes which occur in the CDRs of ABBV-323 Fab upon CD40 binding and 3) develop a structural hypothesis for an agonist/antagonist switch in the LCDR1 of this proprietary class of CD40 antibodies. CONCLUSIONS: The structure of ABBV-323 Fab demonstrates a unique method for antagonism by stabilizing the proposed functional antiparallel dimer for CD40 receptor via novel contacts to LCDR1, namely residue position R32 which is further supported by a closely related agonist antibody FAB516 which shows only monomeric recognition and no contacts with LCDR1 due to a mutation to L32 on LCDR1. These data provide a structural basis for the full antagonist activity of ABBV-323.

Fragment-Based Discovery of an Apolipoprotein E4 (apoE4) Stabilizer.

Apolipoprotein E is a 299-residue lipid carrier protein produced in both the liver and the brain. The protein has three major isoforms denoted apoE2, apoE3, and apoE4 which differ at positions 112 and 158 and which occur at different frequencies in the human population. Genome-wide association studies indicate that the possession of two apoE4 alleles is a strong genetic risk factor for late-onset Alzheimer's disease (LOAD). In an attempt to identify a small molecule stabilizer of apoE4 function that may have utility as a therapy for Alzheimer's disease, we carried out an NMR-based fragment screen on the N-terminal domain of apoE4 and identified a benzyl amidine based fragment binder. In addition to NMR, binding was characterized using various other biophysical techniques, and a crystal structure of the bound core was obtained. Core elaboration ultimately yielded a compound that showed activity in an IL-6 and IL-8 cytokine release assay.

Fragment-based discovery of a potent NAMPT inhibitor.

NAMPT expression is elevated in many cancers, making this protein a potential target for anticancer therapy. We have carried out both NMR based and TR-FRET based fragment screens against human NAMPT and identified six novel binders with a range of potencies. Co-crystal structures were obtained for two of the fragments bound to NAMPT while for the other four fragments force-field driven docking was employed to generate a bound pose. Based on structural insights arising from comparison of the bound fragment poses to that of bound FK866 we were able to synthetically elaborate one of the fragments into a potent NAMPT inhibitor.

Cell-Surface Receptor-Ligand Interaction Analysis with Homogeneous Time-Resolved FRET and Metabolic Glycan Engineering: Application to Transmembrane and GPI-Anchored Receptors.

Ligand-binding assays are the linchpin of drug discovery and medicinal chemistry. Cell-surface receptors and their ligands have traditionally been characterized by radioligand-binding assays, which have low temporal and spatial resolution and entail safety risks. Here, we report a powerful alternative (GlycoFRET), where terbium-labeled fluorescent reporters are irreversibly attached to receptors by metabolic glycan engineering. For the first time, we show time-resolved fluorescence resonance energy transfer between receptor glycans and fluorescently labeled ligands. We describe GlycoFRET for a GPI-anchored receptor, a G-protein-coupled receptor, and a heterodimeric cytokine receptor in living cells with excellent sensitivity and high signal-to-background ratios. In contrast to previously described methods, GlycoFRET does not require genetic engineering or antibodies to label receptors. Given that all cell-surface receptors are glycosylated, we expect that GlycoFRET can be generalized with applications in chemical biology and biotechnology, such as target engagement, receptor pharmacology, and high-throughput screening.

Species differences and molecular determinant of TRPA1 cold sensitivity.

TRPA1 is an ion channel and has been proposed as a thermosensor across species. In invertebrate and ancestral vertebrates such as fly, mosquito, frog, lizard and snakes, TRPA1 serves as a heat receptor, a sensory input utilized for heat avoidance or infrared detection. However, in mammals, whether TRPA1 is a receptor for noxious cold is highly controversial, as channel activation by cold was observed by some groups but disputed by others. Here we attribute the discrepancy to species differences. We show that cold activates rat and mouse TRPA1 but not human or rhesus monkey TRPA1. At the molecular level, a single residue within the S5 transmembrane domain (G878 in rodent but V875 in primate) accounts for the observed difference in cold sensitivity. This residue difference also underlies the species-specific effects of menthol. Together, our findings identify the species-specific cold activation of TRPA1 and reveal a molecular determinant of cold-sensitive gating.

Ligand association rates to the inner-variable-domain of a dual-variable-domain immunoglobulin are significantly impacted by linker design.

The DVD-Ig (TM) protein is a dual-specific immunoglobulin. Each of the two arms of the molecule contains two variable domains, an inner variable domain and an outer variable domain linked in tandem, each with binding specificity for different targets or epitopes. One area of on-going research involves determining how the proximity of the outer variable domain affects the binding of ligands to the inner variable domain. To explore this area, we prepared a series of DVD-Ig proteins with binding specificities toward TNFα and an alternate therapeutic target. Kinetic measurements of TNFα binding to this series of DVD-Ig proteins were used to probe the effects of variable domain position and linker design on ligand on- and off-rates. We found that affinities for TNFα are generally lower when binding to the inner domain than to the outer domain and that this loss of affinity is primarily due to reduced association rate. This effect could be mitigated, to some degree, by linker design. We show several linker sequences that mitigate inner domain affinity losses in this series of DVD-Ig proteins. Moreover, we show that single chain proteolytic cleavage between the inner and outer domains, or complete outer domain removal, can largely restore inner domain TNFα affinity to that approaching the reference antibody. Taken together, these results suggest that a loss of affinity for inner variable domains in this set of DVD-Ig proteins may be largely driven by simple steric hindrance effects and can be reduced by careful linker design.

Development of ELISA to measure TRPV1 protein in rat tissues.

The transient receptor potential vanilloid receptor type 1 (TRPV1) is a non-selective cation channel expressed in both the peripheral and the central nervous systems. To quantitatively determine TRPV1 protein levels in native rat tissues, novel monoclonal antibodies were raised against full-length recombinant human TRPV1 protein and utilized to develop a sandwich ELISA assay. Monoclonal antibody 10E3-1A2 specifically recognized TRPV1 protein and the recognition epitope was determined to reside in amino acids 45-58 of human and rat TRPV1. Using the TRPV1 polyclonal antibody ABRK4 as the capturing antibody and the monoclonal antibody 10E3-1A2 as the detection antibody, a sandwich ELISA that detected both human and rat TRPV1 protein was established. Recombinant human TRPV1 heterologously expressed in mammalian HEK293-F cells, which showed high ligand-binding affinity, was purified by TRPV1 monoclonal antibody affinity chromatography and used as protein standard to quantify TRPV1 protein levels. This ELISA detected TRPV1 protein as low as 1.5ng/ml (15pM), and was able to determine TRPV1 protein levels in native rat tissues such as DRG and spinal cord. This is the first TRPV1 sandwich ELISA that determines the abundance of TRPV1 protein in different tissues. It provides a powerful tool to quantify changes of TRPV1 protein levels in pathological states.

Application of large-scale transient transfection to cell-based functional assays for ion channels and GPCRs.

Despite increasing use of cell-based assays in biomedical research and drug discovery, one challenge is the adequate supply of high-quality cells expressing the target of interest. To this end, stable cell lines expressing the target are often established, maintained, and expanded in large-scale cell culture. These steps require significant investment of time and resources. Moreover, variability occurs regularly in cell yield, viability, expression, and target activities. In particular, stable expression of many targets, such as ion channels, causes toxicity, cell line degeneration, and loss of functional activity. To circumvent these problems, we utilize large-scale transient transfection (LSTT) to generate a large quantity of cells, which are cryopreserved and readily available for use in cell-based functional assays. Here we describe the application of LSTT cells to ion channel and G protein-coupled receptor (GPCR) assays in a drug discovery setting. This approach can also be applied to many other assay formats and target classes.

Expression and purification of human TRPV1 in baculovirus-infected insect cells for structural studies.

TRPV1 is a ligand-gated cation channel that is involved in acute thermal nociception and neurogenic inflammation. By using the GP67 signal peptide, high levels of full-length human TRPV1 was expressed in High Five insect cells using the baculovirus expression system. The functional activity of the expressed TRPV1 was confirmed by whole-cell ligand-gated ion flux recordings in the presence of capsaicin and low pH and via specific ligand binding to the isolated cellular membranes. Efficient solubilization and purification protocols have resulted in milligram amounts of detergent-solubilized channel at 80-90% purity after Ni2+ IMAC chromatography and size exclusion chromatography. Western blot analysis of amino and carboxyl terminal domains and MS of tryptic digestions of purified protein confirmed the presence of the full-length human TRPV1. Specific ligand binding experiments confirmed the protein integrity of the purified human TRPV1.

Solution structure and mutational analysis of pituitary adenylate cyclase-activating polypeptide binding to the extracellular domain of PAC1-RS.

The pituitary adenylate cyclase-activating polypeptide (PACAP) receptor is a class II G protein-coupled receptor that contributes to many different cellular functions including neurotransmission, neuronal survival, and synaptic plasticity. The solution structure of the potent antagonist PACAP (residues 6'-38') complexed to the N-terminal extracellular (EC) domain of the human splice variant hPAC1-R-short (hPAC1-R(S)) was determined by NMR. The PACAP peptide adopts a helical conformation when bound to hPAC1-R(S) with a bend at residue A18' and makes extensive hydrophobic and electrostatic interactions along the exposed beta-sheet and interconnecting loops of the N-terminal EC domain. Mutagenesis data on both the peptide and the receptor delineate the critical interactions between the C terminus of the peptide and the C terminus of the EC domain that define the high affinity and specificity of hormone binding to hPAC1-R(S). These results present a structural basis for hPAC1-R(S) selectivity for PACAP versus the vasoactive intestinal peptide and also differentiate PACAP residues involved in binding to the N-terminal extracellular domain versus other parts of the full-length hPAC1-R(S) receptor. The structural, mutational, and binding data are consistent with a model for peptide binding in which the C terminus of the peptide hormone interacts almost exclusively with the N-terminal EC domain, whereas the central region makes contacts to both the N-terminal and other extracellular parts of the receptor, ultimately positioning the N terminus of the peptide to contact the transmembrane region and result in receptor activation.

Activation of TRPA1 channels by the fatty acid amide hydrolase inhibitor 3'-carbamoylbiphenyl-3-yl cyclohexylcarbamate (URB597).

As a member of the transient receptor potential (TRP) ion channel superfamily, the ligand-gated ion channel TRPA1 has been implicated in nociceptive function and pain states. The endogenous ligands that activate TRPA1 remain unknown. However, various agonists have been identified, including environmental irritants (e.g., acrolein) and ingredients of pungent natural products [e.g., allyl isothiocyanate (ITC), cinnamaldehyde, allicin, and gingerol]. In general, these agents are either highly reactive, nonselective, or not potent or efficacious, significantly limiting their utilities in the study of TRPA1 channel properties and biological functions. In a search for novel TRPA1 agonists, we identified 3'-carbamoylbiphenyl-3-yl cyclohexylcarbamate (URB597), a potent and systemically active inhibitor of fatty acid amide hydrolase (FAAH). This enzyme is responsible for anandamide degradation and therefore has been pursued as an antinociceptive and antiepileptic drug target. Using Ca(2+) influx assays and patch-clamp techniques, we demonstrated that URB597 could activate heterologously expressed human and rat TRPA1 channels, whereas two other FAAH inhibitors (i.e., URB532 and Compound 7) had no effect. When applied to inside-out membrane patches expressing rat TRPA1, URB597 elicited single-channel activities with a unitary conductance of 40 pS. Furthermore, URB597 activated TRPA1 channels endogenously expressed in a population of rat dorsal root ganglion neurons that also responded to ITC. In contrast to its effect on TRPA1, URB597 inhibited TRPM8 and had no effects on TRPV1 or TRPV4. Thus, we conclude that URB597 is a novel agonist of TRPA1 and probably activates the channel through a direct gating mechanism.

Utility of large-scale transiently transfected cells for cell-based high-throughput screens to identify transient receptor potential channel A1 (TRPA1) antagonists.

Despite increasing use of cell-based assays in high-throughput screening (HTS) and lead optimization, one challenge is the adequate supply of high-quality cells expressing the target of interest. To this end, cell lines stably expressing targets are often established, maintained, and scaled up by cell culture. These steps require large investments of time and resources. Moreover, significant variability invariably occurs in cell yield, viability, expression levels, and target activities. In particular, stable expression of targets such as transient receptor potential A1 (TRPA1) causes toxicity, cell line degeneration, and loss of functional activity. Therefore, in an effort to identify TRPA1 antagonists, the authors used large-scale transiently transfected (LSTT) cells, enabling rapid establishment of assays suitable for HTS. LSTT cells, which could- be stored frozen for a long period of time (e.g., at least 42 weeks), retained TRPA1 protein expression and could be easily revived to produce robust and consistent signals in calcium influx and electrophysiological assays. Using cells from a single transfection, a chemical library of 700,000 compounds was screened, and TRPA1 antagonists were identified. The use of LSTT circumvented issues associated with stable TRPA1 expression, increased flexibility and consistency, and greatly reduced labor and cost. This approach will also be applicable to other pharmaceutical targets.

Discovery, structure-activity relationship, and pharmacological evaluation of (5-substituted-pyrrolidinyl-2-carbonyl)-2-cyanopyrrolidines as potent dipeptidyl peptidase IV inhibitors.

A series of (5-substituted pyrrolidinyl-2-carbonyl)-2-cyanopyrrolidine (C5-Pro-Pro) analogues was discovered as dipeptidyl peptidase IV (DPPIV) inhibitors as a potential treatment of diabetes and obesity. X-ray crystallography data show that these inhibitors bind to the catalytic site of DPPIV with the cyano group forming a covalent bond with the serine residue of DPPIV. The C5-substituents make various interactions with the enzyme and affect potency, chemical stability, selectivity, and PK properties of the inhibitors. Optimized analogues are extremely potent with subnanomolar K(i)'s, are chemically stable, show very little potency decrease in the presence of plasma, and exhibit more than 1,000-fold selectivity against related peptidases. The best compounds also possess good PK and are efficacious in lowering blood glucose in an oral glucose tolerance test in ZDF rats.