Concerted Antibody and Antigen Discovery by Differential Whole-cell Phage Display Selections and Multi-omic Target Deconvolution.

Monoclonal antibody (mAb)-based biologics are well established treatments of cancer. Antibody discovery campaigns are typically directed at a single target of interest, which inherently limits the possibility of uncovering novel antibody specificities or functionalities. Here, we present a target-unbiased approach for antibody discovery that relies on generating mAbs against native target cell surfaces via phage display. This method combines a previously reported method for improved whole-cell phage display selections with next-generation sequencing analysis to efficiently identify mAbs with the desired target cell reactivity. Applying this method to multiple myeloma cells yielded a panel of >50 mAbs with unique sequences and diverse reactivities. To uncover the identities of the cognate antigens recognized by this panel, representative mAbs from each unique reactivity cluster were used in a multi-omic target deconvolution approach. From this, we identified and validated three cell surface antigens: PTPRG, ICAM1, and CADM1. PTPRG and CADM1 remain largely unstudied in the context of multiple myeloma, which could warrant further investigation into their potential as therapeutic targets. These results highlight the utility of optimized whole-cell phage display selection methods and could motivate further interest in target-unbiased antibody discovery workflows.

A Sortase A Programmable Phage Display Format for Improved Panning of Fab Antibody Libraries.

Phage display of combinatorial antibody libraries is a versatile tool in the field of antibody engineering, with diverse applications including monoclonal antibody (mAb) discovery, affinity maturation, and humanization. To improve the selection efficiency of antibody libraries, we developed a new phagemid display system that addresses the complication of bald phage propagation. The phagemid facilitates the biotinylation of fragment of antigen binding (Fab) antibody fragments displayed on phage via Sortase A catalysis and the subsequent enrichment of Fab-displaying phage during selections. In multiple contexts, this selection approach improved the enrichment of target-reactive mAbs by depleting background phage. Panels of cancer cell line-reactive mAbs with high diversity and specificity were isolated from a naïve chimeric rabbit/human Fab library using this approach, highlighting its potential to accelerate antibody engineering efforts and to empower concerted antibody drug and target discovery.

Human Serum Albumin Domain I Fusion Protein for Antibody Conjugation.

Bioorthogonal labeling of antibodies enables the conjugation of compounds, such as small molecules or peptides, which expand targeting capacity or enhance cytotoxicity. Taking advantage of a cyclohexene sulfonamide compound that site-selectively labels Lys64 in human serum albumin (HSA), we demonstrate that domain I of HSA can be used as a fusion protein for the preparation of antibody conjugates. Trastuzumab fusions were expressed at the N-terminus of the light chain or the C-terminus of the heavy chain enabling conjugation to small molecules. Moreover, these conjugates retained HER2 binding and proved to be highly stable in human plasma. Antibody conjugation via HSA domain I fusion should therefore have broad utility for making serum-stable antibody conjugates, particularly for antibody-drug conjugates.

Identification of a Binding Site for Unsaturated Fatty Acids in the Orphan Nuclear Receptor Nurr1.

Nurr1/NR4A2 is an orphan nuclear receptor, and currently there are no known natural ligands that bind Nurr1. A recent metabolomics study identified unsaturated fatty acids, including arachidonic acid and docosahexaenoic acid (DHA), that interact with the ligand-binding domain (LBD) of a related orphan receptor, Nur77/NR4A1. However, the binding location and whether these ligands bind other NR4A receptors were not defined. Here, we show that unsaturated fatty acids also interact with the Nurr1 LBD, and solution NMR spectroscopy reveals the binding epitope of DHA at its putative ligand-binding pocket. Biochemical assays reveal that DHA-bound Nurr1 interacts with high affinity with a peptide derived from PIASγ, a protein that interacts with Nurr1 in cellular extracts, and DHA also affects cellular Nurr1 transactivation. This work is the first structural report of a natural ligand binding to a canonical NR4A ligand-binding pocket and indicates a natural ligand can bind and affect Nurr1 function.

Deconvolution of Complex 1D NMR Spectra Using Objective Model Selection.

Fluorine (19F) NMR has emerged as a useful tool for characterization of slow dynamics in 19F-labeled proteins. One-dimensional (1D) 19F NMR spectra of proteins can be broad, irregular and complex, due to exchange of probe nuclei between distinct electrostatic environments; and therefore cannot be deconvoluted and analyzed in an objective way using currently available software. We have developed a Python-based deconvolution program, decon1d, which uses Bayesian information criteria (BIC) to objectively determine which model (number of peaks) would most likely produce the experimentally obtained data. The method also allows for fitting of intermediate exchange spectra, which is not supported by current software in the absence of a specific kinetic model. In current methods, determination of the deconvolution model best supported by the data is done manually through comparison of residual error values, which can be time consuming and requires model selection by the user. In contrast, the BIC method used by decond1d provides a quantitative method for model comparison that penalizes for model complexity helping to prevent over-fitting of the data and allows identification of the most parsimonious model. The decon1d program is freely available as a downloadable Python script at the project website (https://github.com/hughests/decon1d/).

Inhibition of Non-ATG Translational Events in Cells via Covalent Small Molecules Targeting RNA.

One major class of disease-causing RNAs is expanded repeating transcripts. These RNAs cause diseases via multiple mechanisms, including: (i) gain-of-function, in which repeating RNAs bind and sequester proteins involved in RNA biogenesis and (ii) repeat associated non-ATG (RAN) translation, in which repeating transcripts are translated into toxic proteins without use of a canonical, AUG, start codon. Herein, we develop and study chemical probes that bind and react with an expanded r(CGG) repeat (r(CGG)(exp)) present in a 5' untranslated region that causes fragile X-associated tremor/ataxia syndrome (FXTAS). Reactive compounds bind to r(CGG)(exp) in cellulo as shown with Chem-CLIP-Map, an approach to map small molecule binding sites within RNAs in cells. Compounds also potently improve FXTAS-associated pre-mRNA splicing and RAN translational defects, while not affecting translation of the downstream open reading frame. In contrast, oligonucleotides affect both RAN and canonical translation when they bind to r(CGG)(exp), which is mechanistically traced to a decrease in polysome loading. Thus, designer small molecules that react with RNA targets can be used to profile the RNAs to which they bind in cells, including identification of binding sites, and can modulate several aspects of RNA-mediated disease pathology in a manner that may be more beneficial than oligonucleotides.

Tethered particle motion reveals that LacI·DNA loops coexist with a competitor-resistant but apparently unlooped conformation.

The lac repressor protein (LacI) efficiently represses transcription of the lac operon in Escherichia coli by binding to two distant operator sites on the bacterial DNA and causing the intervening DNA to form a loop. We employed single-molecule tethered particle motion to observe LacI-mediated loop formation and breakdown in DNA constructs that incorporate optimized operator binding sites and intrinsic curvature favorable to loop formation. Previous bulk competition assays indirectly measured the loop lifetimes in these optimized DNA constructs as being on the order of days; however, we measured these same lifetimes to be on the order of minutes for both looped and unlooped states. In a range of single-molecule DNA competition experiments, we found that the resistance of the LacI-DNA complex to competitive binding is a function of both the operator strength and the interoperator sequence. To explain these findings, we present what we believe to be a new kinetic model of loop formation and DNA competition. In this proposed new model, we hypothesize a new unlooped state in which the unbound DNA-binding domain of the LacI protein interacts nonspecifically with nonoperator DNA adjacent to the operator site at which the second LacI DNA-binding domain is bound.