A comparison of the binding sites of antibodies and single-domain antibodies.

Antibodies are the largest class of biotherapeutics. However, in recent years, single-domain antibodies have gained traction due to their smaller size and comparable binding affinity. Antibodies (Abs) and single-domain antibodies (sdAbs) differ in the structures of their binding sites: most significantly, single-domain antibodies lack a light chain and so have just three CDR loops. Given this inherent structural difference, it is important to understand whether Abs and sdAbs are distinguishable in how they engage a binding partner and thus, whether they are suited to different types of epitopes. In this study, we use non-redundant sequence and structural datasets to compare the paratopes, epitopes and antigen interactions of Abs and sdAbs. We demonstrate that even though sdAbs have smaller paratopes, they target epitopes of equal size to those targeted by Abs. To achieve this, the paratopes of sdAbs contribute more interactions per residue than the paratopes of Abs. Additionally, we find that conserved framework residues are of increased importance in the paratopes of sdAbs, suggesting that they include non-specific interactions to achieve comparable affinity. Furthermore, the epitopes of sdAbs are only marginally less accessible than those of Abs: we posit that this may be explained by differences in the orientation and compaction of sdAb and Ab CDR-H3 loops. Overall, our results have important implications for the engineering and humanization of sdAbs, as well as the selection of the best modality for targeting a particular epitope.

Discovery of STRO-002, a Novel Homogeneous ADC Targeting Folate Receptor Alpha, for the Treatment of Ovarian and Endometrial Cancers.

STRO-002 is a novel homogeneous folate receptor alpha (FolRα) targeting antibody-drug conjugate (ADC) currently being investigated in the clinic as a treatment for ovarian and endometrial cancers. Here, we describe the discovery, optimization, and antitumor properties of STRO-002. STRO-002 was generated by conjugation of a novel cleavable 3-aminophenyl hemiasterlin linker-warhead (SC239) to the nonnatural amino acid para-azidomethyl-L-phenylalanine incorporated at specific positions within a high affinity anti-FolRα antibody using Sutro's XpressCF+, which resulted in a homogeneous ADC with a drug-antibody ratio (DAR) of 4. STRO-002 binds to FolRα with high affinity, internalizes rapidly into target positive cells, and releases the tubulin-targeting cytotoxin 3-aminophenyl hemiasterlin (SC209). SC209 has reduced potential for drug efflux via P-glycoprotein 1 drug pump compared with other tubulin-targeting payloads. While STRO-002 lacks nonspecific cytotoxicity toward FolRα-negative cell lines, bystander killing of target negative cells was observed when cocultured with target positive cells. STRO-002 is stable in circulation with no change in DAR for up to 21 days and has a half-life of 6.4 days in mice. A single dose of STRO-002 induced significant tumor growth inhibition in FolRα-expressing xenograft models and patient-derived xenograft models. In addition, combination treatment with carboplatin or Avastin further increased STRO-002 efficacy in xenograft models. The potent and specific preclinical efficacy of STRO-002 supports clinical development of STRO-002 for treating patients with FolRα-expressing cancers, including ovarian, endometrial, and non-small cell lung cancer. Phase I dose escalation for STRO-002 is in progress in ovarian cancer and endometrial cancer patients (NCT03748186 and NCT05200364).

RF1 attenuation enables efficient non-natural amino acid incorporation for production of homogeneous antibody drug conjugates.

Amber codon suppression for the insertion of non-natural amino acids (nnAAs) is limited by competition with release factor 1 (RF1). Here we describe the genome engineering of a RF1 mutant strain that enhances suppression efficiency during cell-free protein synthesis, without significantly impacting cell growth during biomass production. Specifically, an out membrane protease (OmpT) cleavage site was engineered into the switch loop of RF1, which enables its conditional inactivation during cell lysis. This facilitates extract production without additional processing steps, resulting in a scaleable extract production process. The RF1 mutant extract allows nnAA incorporation at previously intractable sites of an IgG1 and at multiple sites in the same polypeptide chain. Conjugation of cytotoxic agents to these nnAAs, yields homogeneous antibody drug conjugates (ADCs) that can be optimized for conjugation site, drug to antibody ratio (DAR) and linker-warheads designed for efficient tumor killing. This platform provides the means to generate therapeutic ADCs inaccessible by other methods that are efficient in their cytotoxin delivery to tumor with reduced dose-limiting toxicities and thus have the potential for better clinical impact.

Production of bispecific antibodies in "knobs-into-holes" using a cell-free expression system.

Bispecific antibodies have emerged in recent years as a promising field of research for therapies in oncology, inflammable diseases, and infectious diseases. Their capability of dual target recognition allows for novel therapeutic hypothesis to be tested, where traditional mono-specific antibodies would lack the needed mode of target engagement. Among extremely diverse architectures of bispecific antibodies, knobs-into-holes (KIHs) technology, which involves engineering CH3 domains to create either a "knob" or a "hole" in each heavy chain to promote heterodimerization, has been widely applied. Here, we describe the use of a cell-free expression system (Xpress CF) to produce KIH bispecific antibodies in multiple scaffolds, including 2-armed heterodimeric scFv-KIH and one-armed asymmetric BiTE-KIH with tandem scFv. Efficient KIH production can be achieved by manipulating the plasmid ratio between knob and hole, and further improved by addition of prefabricated knob or hole. These studies demonstrate the versatility of Xpress CF in KIH production and provide valuable insights into KIH construct design for better assembly and expression titer.

A general sequence processing and analysis program for protein engineering.

Protein engineering projects often amass numerous raw DNA sequences, but no readily available software combines sequence processing and activity correlation required for efficient lead identification. XLibraryDisplay is an open source program integrated into Microsoft Excel for Windows that automates batch sequence processing via a simple step-by-step, menu-driven graphical user interface. XLibraryDisplay accepts any DNA template which is used as a basis for trimming, filtering, translating, and aligning hundreds to thousands of sequences (raw, FASTA, or Phred PHD file formats). Key steps for library characterization through lead discovery are available including library composition analysis, filtering by experimental data, graphing and correlating to experimental data, alignment to structural data extracted from PDB files, and generation of PyMOL visualization scripts. Though larger data sets can be handled, the program is best suited for analyzing approximately 10 000 or fewer leads or naïve clones which have been characterized using Sanger sequencing and other experimental approaches. XLibraryDisplay can be downloaded for free from sourceforge.net/projects/xlibrarydisplay/ .

In vitro Fab display: a cell-free system for IgG discovery.

Selection technologies such as ribosome display enable the rapid discovery of novel antibody fragments entirely in vitro. It has been assumed that the open nature of the cell-free reactions used in these technologies limits selections to single-chain protein fragments. We present a simple approach for the selection of multi-chain proteins, such as antibody Fab fragments, using ribosome display. Specifically, we show that a two-chain trastuzumab (Herceptin) Fab domain can be displayed in a format which tethers either the heavy or light chain to the ribosome while retaining functional antigen binding. Then, we constructed synthetic Fab HC and LC libraries and performed test selections against carcinoembryonic antigen (CEA) and vascular endothelial growth factor (VEGF). The Fab selection output was reformatted into full-length immunoglobulin Gs (IgGs) and directly expressed at high levels in an optimized cell-free system for immediate screening, purification and characterization. Several novel IgGs were identified using this cell-free platform that bind to purified CEA, CEA positive cells and VEGF.

Tandem SAM domain structure of human Caskin1: a presynaptic, self-assembling scaffold for CASK.

The synaptic scaffolding proteins CASK and Caskin1 are part of the fibrous mesh of proteins that organize the active zones of neural synapses. CASK binds to a region of Caskin1 called the CASK interaction domain (CID). Adjacent to the CID, Caskin1 contains two tandem sterile α motif (SAM) domains. Many SAM domains form polymers so they are good candidates for forming the fibrous structures seen in the active zone. We show here that the SAM domains of Caskin1 form a new type of SAM helical polymer. The Caskin1 polymer interface exhibits a remarkable segregation of charged residues, resulting in a high sensitivity to ionic strength in vitro. The Caskin1 polymers can be decorated with CASK proteins, illustrating how these proteins may work together to organize the cytomatrix in active zones.

The molecular basis of the Caskin1 and Mint1 interaction with CASK.

Calcium/calmodulin-dependent serine protein kinase (CASK) is a conserved multi-domain scaffolding protein involved in brain development, synapse formation, and establishment of cell polarity. To accomplish these diverse functions, CASK participates in numerous protein-protein interactions. In particular, CASK forms competing CASK/Mint1/Velis and CASK/Caskin1/Velis tripartite complexes that physically associate with the cytoplasmic tail of neurexin, a transmembrane protein enriched at presynaptic sites. This study shows that a short linear EEIWVLRK peptide motif from Caskin1 is necessary and sufficient for binding CASK. We also identified the conserved binding site for the peptide on the CASK calmodulin kinase domain. A related EPIWVMRQ peptide from Mint1 was also discovered to be sufficient for binding. Searching all human proteins for the Mint1/Caskin1 consensus peptide ExIWVxR revealed that T-cell lymphoma invasion and metastasis 1 (TIAM1) contains a conserved EEVIWVRRE peptide that was also found to be sufficient for CASK binding in vitro. TIAM1 is well known for its role in tumor metastasis, but it also possesses overlapping cellular and neurological functions with CASK, suggesting a previously unknown cooperation between the two proteins. This new peptide interaction motif also explains how Caskin1 and Mint1 form competing complexes and suggests a new role for the cellular hub protein CASK.

Crystal structure of the central coiled-coil domain from human liprin-ß2.

Liprins are a conserved family of scaffolding proteins important for the proper regulation and development of neuronal synapses. Humans have four liprin-αs and two liprin-ßs which all contain long coiled-coil domains followed by three tandem SAM domains. Complex interactions between the coiled-coil and SAM domains are thought to create liprin scaffolds, but the structural and biochemical properties of these domains remain largely uncharacterized. In this study we find that the human liprin-ß2 coiled-coil forms an extended dimer. Several protease-resistant subdomains within the liprin-ß1 and liprin-ß2 coiled-coils were also identified. A 2.0 Å crystal structure of the central, protease-resistant core of the liprin-ß2 coiled-coil reveals a parallel helix orientation. These studies represent an initial step toward determining the overall architecture of liprin scaffolds and understanding the molecular basis for their synaptic functions.

Targeted chemical wedges reveal the role of allosteric DNA modulation in protein-DNA assembly.

The cooperative assembly of multiprotein complexes results from allosteric modulations of DNA structure as well as direct intermolecular contacts between proteins. Such cooperative binding plays a critical role in imparting exquisite sequence specificity on the homeobox transcription factor (Hox) family of developmental transcription factors. A well-characterized example includes the interaction of Hox proteins with extradenticle (Exd), a highly conserved DNA binding transcription factor. Although direct interactions are important, the contribution of indirect interactions toward cooperative assembly of Hox and Exd remains unresolved. Here we use minor groove binding polyamides as structural wedges to induce perturbations at specific base steps within the Exd binding site. We find that allosteric modulation of DNA structure contributes nearly 1.5 kcal/mol to the binding of Exd to DNA, even in the absence of direct Hox contacts. In contrast to previous studies, the sequence-targeted chemical wedges reveal the role of DNA geometry in cooperative assembly of Hox-Exd complexes. Programmable polyamides may well serve as general probes to investigate the role of DNA modulation in the cooperative and highly specific assembly of other protein-DNA complexes.

The reach of linear protein-DNA dimerizers.

A protein-DNA dimerizer constructed from a DNA-binding pyrrole-imidazole polyamide and the peptide FYPWMK facilitates binding of the natural transcription factor Exd to an adjacent DNA site. Previous dimerizers have been constructed with the peptide attached to an internal pyrrole monomer in an overall branched oligomer. Linear oligomers constructed by attaching the peptide to the polyamide C-terminus expand the range of protein-DNA dimerization to six additional DNA sites. Replacing the FYPWMK hexapeptide with a WM dipeptide, which was previously functional in branched compounds, does not lead to a functional linear dimerizer. Instead, inserting an additional lysine generates a minimal, linear WMK tripeptide conjugate that maintains the activity of the larger FYPWMK dimerizers in a single DNA-binding site orientation. These studies provide insight into the importance of linker length and composition, binding site spacing and orientation, and the protein-binding domain content that are important for the optimization of protein-DNA dimerizers suitable for biological experiments.

Minimization of a protein-DNA dimerizer.

A protein-DNA dimerizer constructed from a DNA-binding polyamide and the peptide FYPWMKG facilitates the binding of a natural transcription factor Exd to an adjacent DNA site. The Exd binding domain can be reduced to a dipeptide WM attached to the polyamide through an epsilon-aminohexanoic acid linker with retention of protein-DNA dimerizer activity. Screening a library of analogues indicated that the tryptophan indole moiety is more important than methionine's side chain or the N-terminal acetamide. Remarkably, switching the stereochemistry of the tryptophan residue (l to d) stabilizes the dimerizer*Exd*DNA ternary complex at 37 degrees C. These observations provide design principles for artificial transcription factors that may function in concert with the cellular regulatory circuitry.

EF-Tu binding peptides identified, dissected, and affinity optimized by phage display.

The highly abundant GTP binding protein elongation factor Tu (EF-Tu) fulfills multiple roles in bacterial protein biosynthesis. Phage-displayed peptides with high affinity for EF-Tu were selected from a library of approximately 4.7 x 10(11) different peptides. The lack of sequence homology among the identified EF-Tu ligands demonstrates promiscuous peptide binding by EF-Tu. Homolog shotgun scanning of an EF-Tu ligand was used to dissect peptide molecular recognition by EF-Tu. All homolog shotgun scanning selectants bound to EF-Tu with higher affinity than the starting ligand. Thus, homolog shotgun scanning can simultaneously optimize binding affinity and rapidly provide detailed structure activity relationships for multiple side chains of a polypeptide ligand. The reported peptide ligands do not compete for binding to EF-Tu with various antibiotic EF-Tu inhibitors, and could identify an EF-Tu peptide binding site distinct from the antibiotic inhibitory sites.

Dissecting the streptavidin-biotin interaction by phage-displayed shotgun scanning.

Shotgun scanning the streptavidin-biotin interaction identifies long-range hydrophobic interactions that contribute to one of the strongest naturally occurring noncovalent protein-ligand interactions. The femtomolar dissociation constant for this interaction makes it a useful model system to dissect the forces that govern high-affinity molecular recognition between proteins and small molecules. Shotgun scanning combines the diversity and in vitro binding selection of phage-displayed libraries with a binomial mutagenesis strategy. Libraries consist of proteins with the residues in multiple positions mutated to give a 1:1 ratio of alanine:wild type. Here, we use shotgun scanning to determine the functional contribution of the 38 C-terminal residues of streptavidin to the high-affinity interaction with biotin. The library pools were subjected to three rounds of selection for functional streptavidin variants that bind biotin and statistical analysis was used to assess side-chain contributions to biotin binding. The results demonstrate the utility of shotgun scanning for the dissection of receptor-small-molecule interactions. While shotgun scanning results were largely consistent with previous single-point, site-directed mutagenesis studies for residues in direct contact with biotin, residues distant from the biotin binding site have not previously been explored. Key streptavidin residues identified by shotgun scanning as contributors to the interaction with biotin include those with side chains that fill the beta barrel, residues at the tetramer interface, and second-sphere residues, which are reinforced by long-distance propagation of hydrophobic interactions.