Engineering CD276/B7-H3-targeted antibody-drug conjugates with enhanced cancer-eradicating capability

CD276/B7-H3 represents a promising target for cancer therapy based on widespread overexpression in both cancer cells and tumor-associated stroma. In previous preclinical studies, CD276 antibody-drug conjugates (ADCs) exploiting a talirine-type pyrrolobenzodiazepine (PBD) payload showed potent activity against various solid tumors but with a narrow therapeutic index and dosing regimen higher than that tolerated in clinical trials using other antibody-talirine conjugates. Here, we describe the development of a modified talirine PBD-based fully human CD276 ADC, called m276-SL-PBD, that is cross-species (human/mouse) reactive and can eradicate large 500-1,000-mm3 triple-negative breast cancer xenografts at doses 10- to 40-fold lower than the maximum tolerated dose. By combining CD276 targeting with judicious genetic and chemical ADC engineering, improved ADC purification, and payload sensitivity screening, these studies demonstrate that the therapeutic index of ADCs can be substantially increased, providing an advanced ADC development platform for potent and selective targeting of multiple solid tumor types.

An Efficient Opal-Suppressor Tryptophanyl Pair Creates New Routes for Simultaneously Incorporating up to Three Distinct Noncanonical Amino Acids into Proteins in Mammalian Cells

Site-specific incorporation of multiple distinct noncanonical amino acids (ncAAs) into proteins in mammalian cells is a promising technology, where each ncAA must be assigned to a different orthogonal aminoacyl-tRNA synthetase (aaRS)/tRNA pair that reads a distinct nonsense codon. Available pairs suppress TGA or TAA codons at a considerably lower efficiency than TAG, limiting the scope of this technology. Here we show that the E. coli tryptophanyl (EcTrp) pair is an excellent TGA-suppressor in mammalian cells, which can be combined with the three other established pairs to develop three new routes for dual-ncAA incorporation. Using these platforms, we site-specifically incorporated two different bioconjugation handles into an antibody with excellent efficiency, and subsequently labeled it with two distinct cytotoxic payloads. Additionally, we combined the EcTrp pair with other pairs to site-specifically incorporate three distinct ncAAs into a reporter protein in mammalian cells.

Abstract 1521: Unlocking the therapeutic index of highly potent site-specific ADCs in solid tumors

CD276 (B7-H3) is a transmembrane protein of the B7 family, which includes many prominent immune checkpoint targets, and is expected to play a role in T-cell stimulation, tumor-infiltration, as well as other tumorigenic functions with poor prognosis when upregulated. CD276 is highly expressed in solid tumors, cancer initiating stem cells, and tumor associated vasculature with minimal to no expression in healthy tissue. This promising expression profile has piqued interest for various targeted therapeutic modalities, such as antibody-drug conjugates (ADCs). Developing effective ADCs for solid tumors remains a challenge due to inefficient conjugation technologies which are either non-homogeneous, unstable, or do not position payloads at optimal sites. BrickADCs minimize hydrophobicity, optimize payload release kinetics, and tune biophysical properties to improve efficacy, safety, and bioavailability, resulting in improved therapeutic indexes. Potent payloads such as pyrrolobenzodiazepines have been underutilized due to toxicity, preventing their widespread use. BrickBio’s unique bioconjugation methodology enables precise (site-specific), flexible (unfettered site accessibility and chemistry), efficient, and scalable generation of ADCs which overcome these limitations. In this work, UAAs (unnatural amino acids) were incorporated into full-length anti-CD276 antibodies and conjugated with optimized pyrrolobenzodiazepine (PBD) payloads. BrickADCs leverage a proprietary Site-Select Panel to find optimal sites that shield the toxicity of highly potent, hydrophobic PBD payloads while maintaining their high activity to generate anti-CD276-UAA-PBD ADCs. The highest performing BrickADCs, developed by a rapid screen of PBD payloads conjugated to different sites in the antibody backbone, exhibit high therapeutic efficacy in a variety of CD276 positive solid tumor cell lines (eg. colon, ovarian, lung, sarcoma, etc) and in mouse xenograft models with a 3x differential in complete response. The tuned hydrophobicity and release kinetics have resulted in ADCs that have an increased therapeutic index, particularly due to their improved safety profile compared to cysteine conjugation.The BrickADC platform enables an increased therapeutic window for potent payloads across solid tumors, opening the floodgates for further development of ADCs which leverage payloads that have been plagued by poor therapeutic index. Continued work is underway for the CD276 pipeline candidates as they progress towards IND, as well as investigation into other targets and conjugate modalities.

Citation Format: James S. Italia, Colby A. Souders, Sreedhar Reddy Suthe, Myer Hussain, Zhi Li, John Boyce, Audrey Warner, Abhishek Chatterjee. Unlocking the therapeutic index of highly potent site-specific ADCs in solid tumors [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 1521.

Selective incorporation of 5-hydroxytryptophan blocks long range electron transfer in oxalate decarboxylase

Oxalate decarboxylase from Bacillus subtilis is a binuclear Mn-dependent acid stress response enzyme that converts the mono-anion of oxalic acid into formate and carbon dioxide in a redox neutral unimolecular disproportionation reaction. A π-stacked tryptophan dimer, W96 and W274, at the interface between two monomer subunits facilitates long-range electron transfer between the two Mn ions and plays an important role in the catalytic mechanism. Substitution of W96 with the unnatural amino acid 5-hydroxytryptophan leads to a persistent EPR signal which can be traced back to the neutral radical of 5-hydroxytryptophan with its hydroxyl proton removed. 5-Hydroxytryptophan acts as a hole sink preventing the formation of Mn(III) at the N-terminal active site and strongly suppresses enzymatic activity. The lower boundary of the standard reduction potential for the active site Mn(II)/Mn(III) couple can therefore be estimated as 740 mV against the normal hydrogen electrode at pH 4, the pH of maximum catalytic efficiency. Our results support the catalytic importance of long-range electron transfer in oxalate decarboxylase while at the same time highlighting the utility of unnatural amino acid incorporation and specifically the use of 5-hydroxytryptophan as an energetic sink for hole hopping to probe electron transfer in redox proteins.

A Robust Platform for Unnatural Amino Acid Mutagenesis in E. coli Using the Bacterial Tryptophanyl-tRNA synthetase/tRNA pair

We report the development of a robust user-friendly Escherichia coli (E. coli) expression system, derived from the BL21(DE3) strain, for site-specifically incorporating unnatural amino acids (UAAs) into proteins using engineered E. coli tryptophanyl-tRNA synthetase (EcTrpRS)-tRNA^Trp pairs. This was made possible by functionally replacing the endogenous EcTrpRS-tRNA^Trp pair in BL21(DE3) E. coli with an orthogonal counterpart from Saccharomyces cerevisiae, and reintroducing it into the resulting altered translational machinery tryptophanyl (ATMW-BL21) E. coli strain as an orthogonal nonsense suppressor. The resulting expression system benefits from the favorable characteristics of BL21(DE3) as an expression host, and is compatible with the broadly used T7-driven recombinant expression system. Furthermore, the vector expressing the nonsense-suppressing engineered EcTrpRS-tRNA^Trp pair was systematically optimized to significantly enhance the incorporation efficiency of various tryptophan analogs. Together, the improved strain and the optimized suppressor plasmids enable efficient UAA incorporation (up to 65% of wild-type levels) into several different proteins. This robust and user-friendly platform will significantly expand the scope of the genetically encoded tryptophan-derived UAAs.

Structural Robustness Affects the Engineerability of Aminoacyl-tRNA Synthetases for Genetic Code Expansion

The ability to engineer the substrate specificity of natural aminoacyl-tRNA synthetase/tRNA pairs facilitates the site-specific incorporation of noncanonical amino acids (ncAAs) into proteins. The Methanocaldococcus jannaschii-derived tyrosyl-tRNA synthetase (MjTyrRS)/tRNA pair has been engineered to incorporate numerous ncAAs into protein expressed in bacteria. However, it cannot be used in eukaryotic cells due to cross-reactivity with its host counterparts. The Escherichia coli-derived tyrosyl-tRNA synthetase (EcTyrRS)/tRNA pair offers a suitable alternative to this end, but a much smaller subset of ncAAs have been genetically encoded using this pair. Here we report that this discrepancy, at least partly, stems from the structural robustness of EcTyrRS being lower than that of MjTyrRS. We show that the thermostability of engineered TyrRS mutants is generally significantly lower than those of their wild-type counterparts. Derived from a thermophilic archaeon, MjTyrRS is a remarkably sturdy protein and tolerates extensive active site engineering without a catastrophic loss of stability at physiological temperature. In contrast, EcTyrRS exhibits significantly lower thermostability, rendering some of its engineered mutants insufficiently stable at physiological temperature. Our observations identify the structural robustness of an aaRS as an important factor that significantly influences how extensively it can be engineered. To overcome this limitation, we have further developed chimeras between EcTyrRS and its homologue from a thermophilic bacterium, which offer an optimal balance between thermostability and activity. We show that the chimeric bacterial TyrRSs show enhanced tolerance for destabilizing active site mutations, providing a potentially more engineerable platform for genetic code expansion.

Genetically encoded protein sulfation in mammalian cells

Tyrosine sulfation is an important post-translational modification found in higher eukaryotes. Here we report an engineered tyrosyl-tRNA synthetase/tRNA pair that co-translationally incorporates O-sulfotyrosine in response to UAG codons in E. coli and mammalian cells. This platform enables recombinant expression of eukaryotic proteins homogeneously sulfated at chosen sites, which was demonstrated by expressing human heparin cofactor II in mammalian cells in different states of sulfation.

A "Quenchergenic" Chemoselective Protein Labeling Strategy

Dynamic changes in protein structure can be monitored by using a fluorescent probe and a dark quencher. This approach is contingent upon the ability to precisely introduce a fluorophore/quencher pair into two specific sites of a protein of interest. Despite recent advances, there is continued demand for new and convenient approaches to site-selectively label proteins with such optical probes. We have recently developed a chemoselectively rapid azo-coupling reaction (CRACR) for site-specific protein labeling; it relies on rapid coupling between a genetically encoded 5-hydroxytryptophan residue and various aromatic diazonium ions. Herein, it is reported that the product of this conjugation reaction, a highly chromophoric biarylazo group, is a potent fluorescence quencher. The absorption properties of this azo product can be tuned by systematically altering the structure of the aryldiazonium species. A particular "quenchergenic" aryldiazonium has been identified that, upon conjugation, efficiently quenches the fluorescence of green fluorescent protein, which is a widely used genetically encoded fluorescent probe that can be terminally attached to target proteins. This fluorophore/quencher pair was used to evaluate the protein-labeling kinetics of CRACR, as well as to monitor the proteolysis of a fusion protein.

Synthesis and characterization of an unnatural boron and nitrogen-containing tryptophan analogue and its incorporation into proteins

A boron and nitrogen containing unnatural analogue of tryptophan is synthesized and incorporated into proteins.

A boron and nitrogen containing unnatural analogue of tryptophan is synthesized through the functionalization of BN-indole. The spectroscopic properties of BN-tryptophan are reported with respect to the natural tryptophan, and the incorporation of BN-tryptophan into proteins expressed in E. coli using selective pressure incorporation is described. This work shows that a cellular system can recognize the unnatural, BN-containing tryptophan. More importantly, it presents the first example of an azaborine containing amino acid being incorporated into proteins.

Mutually Orthogonal Nonsense-Suppression Systems and Conjugation Chemistries for Precise Protein Labeling at up to Three Distinct Sites

Site-specific incorporation of multiple distinct noncanonical amino acids (ncAAs) into a protein is an emerging technology with tremendous potential. It relies on mutually orthogonal engineered aminoacyl-tRNA synthetase/tRNA pairs that suppress different nonsense/frameshift codons. So far, up to two distinct ncAAs have been incorporated into proteins expressed in E. coli, using archaea-derived tyrosyl and pyrrolysyl pairs. Here we report that the E. coli derived tryptophanyl pair can be combined with the archaeal tyrosyl or the pyrrolysyl pair in ATMW1 E. coli to incorporate two different ncAAs into one protein with high fidelity and efficiency. By combining all three orthogonal pairs, we further demonstrate simultaneous site-specific incorporation of three different ncAAs into one protein. To use this technology for chemoselectively labeling proteins with multiple distinct entities at predefined sites, we also sought to identify different bioconjugation handles that can be coincorporated into proteins as ncAA-side chains and subsequently functionalized through mutually compatible labeling chemistries. To this end, we show that the recently developed chemoselective rapid azo-coupling reaction (CRACR) directed to 5-hydroxytryptophan (5HTP) is compatible with strain-promoted azide-alkyne cycloaddition (SPAAC) targeted to p-azidophenylalanine (pAzF) and strain-promoted inverse electron-demand Diels-Alder cycloaddition (SPIEDAC) targeted to cyclopropene-lysine (CpK) for rapid, catalyst-free protein labeling at multiple sites. Combining these mutually orthogonal nonsense suppression systems and the mutually compatible bioconjugation handles they incorporate, we demonstrate site-specific labeling of recombinantly expressed proteins at up to three distinct sites.

Labeling Proteins at Site-Specifically Incorporated 5-Hydroxytryptophan Residues Using a Chemoselective Rapid Azo-Coupling Reaction

Chemoselective protein labeling is a valuable tool in the arsenal of modern chemical biology. The unnatural amino acid mutagenesis technology provides a powerful way to site-specifically introduce nonnatural chemical functionalities into recombinant proteins, which can be subsequently functionalized in a chemoselective manner. Even though several strategies currently exist to selectively label recombinant proteins in this manner, there is considerable interest for the development of additional chemoselective reactions that are fast, catalyst-free, use readily available reagents, and are compatible with existing conjugation chemistries. Here we describe a method to express recombinant proteins in E. coli site-specifically incorporating 5-hydroxytryptophan, followed by the chemoselective labeling of this residue using a chemoselective rapid azo-coupling reaction.

Resurrecting the Bacterial Tyrosyl-tRNA Synthetase/tRNA Pair for Expanding the Genetic Code of Both E. coli and Eukaryotes

The bacteria-derived tyrosyl-tRNA synthetase (TyrRS)/tRNA pair was first used for unnatural amino acid (Uaa) mutagenesis in eukaryotic cells over 15 years ago. It provides an ideal platform to genetically encode numerous useful Uaas in eukaryotes. However, this pair has been engineered to charge only a small collection of Uaas to date. Development of Uaa-selective variants of this pair has been limited by technical challenges associated with a yeast-based directed evolution platform, which is currently required to alter its substrate specificity. Here we overcome this limitation by enabling its directed evolution in an engineered strain of E. coli (ATMY), where the endogenous TyrRS/tRNA pair has been functionally replaced with an archaeal counterpart. The facile E. coli-based selection system enabled rapid engineering of this pair to develop variants that selectively incorporate various Uaas, including p-boronophenylalanine, into proteins expressed in mammalian cells as well as in the ATMY strain of E. coli.

An Oxidative Bioconjugation Strategy Targeted to a Genetically Encoded 5-Hydroxytryptophan

Approaches that enable the chemoselective, covalent modification of proteins in a site-specific manner have emerged as a powerful technology for a wide range of applications. The electron-rich unnatural amino acid 5-hydroxytryptophan was recently genetically encoded in both Escherichia coli and eukaryotes, thereby allowing its site-specific incorporation into virtually any recombinant protein. Herein, we report the chemoselective conjugation of various aromatic amines to full-length proteins under mild, oxidative conditions that target this site-specifically incorporated 5-hydroxytryptophan residue.

A Chemoselective Rapid Azo-Coupling Reaction (CRACR) for Unclickable Bioconjugation

Chemoselective modification of complex biomolecules has become a cornerstone of chemical biology. Despite the exciting developments of the past two decades, the demand for new chemoselective reactions with unique abilities, and those compatible with existing chemistries for concurrent multisite-directed labeling, remains high. Here we show that 5-hydroxyindoles exhibit remarkably high reactivity toward aromatic diazonium ions and this reaction can be used to chemoselectively label proteins. We have previously genetically encoded the noncanonical amino acid 5-hydroxytryptophan in both E. coli and eukaryotes, enabling efficient site-specific incorporation of 5-hydroxyindole into virtually any protein. The 5-hydroxytryptophan residue was shown to allow rapid, chemoselective protein modification using the azo-coupling reaction, and the utility of this bioconjugation strategy was further illustrated by generating a functional antibody-fluorophore conjugate. Although the resulting azo-linkage is otherwise stable, we show that it can be efficiently cleaved upon treatment with dithionite. Our work establishes a unique chemoselective "unclickable" bioconjugation strategy to site-specifically modify proteins expressed in both bacteria and eukaryotes.

Performance of optimized noncanonical amino acid mutagenesis systems in the absence of release factor 1

Site-specific incorporation of noncanonical amino acids (ncAAs) into proteins expressed in E. coli using UAG-suppression competes with termination mediated by release factor 1 (RF1). Recently, unconditional deletion of RF1 was achieved in a genomically recoded E. coli (C321), devoid of all endogenous UAG stop codons. Here we evaluate the efficiency of ncAA incorporation in this strain using optimized suppression vectors. Even though the absence of RF1 does not benefit the suppression efficiency of a single UAG codon, multi-site incorporation of a series of chemically distinct ncAAs was significantly improved.

An orthogonalized platform for genetic code expansion in both bacteria and eukaryotes

In this study, we demonstrate the feasibility of expanding the genetic code of Escherichia coli using its own tryptophanyl-tRNA synthetase and tRNA (TrpRS-tRNA^Trp) pair. This was made possible by first functionally replacing this endogenous pair with an E. coli-optimized counterpart from Saccharomyces cerevisiae, and then reintroducing the liberated E. coli TrpRS-tRNA^Trp pair into the resulting strain as a nonsense suppressor, which was then followed by its directed evolution to genetically encode several new unnatural amino acids (UAAs). These engineered TrpRS-tRNA^Trp variants were also able to drive efficient UAA mutagenesis in mammalian cells. Since bacteria-derived aminoacyl-tRNA synthetase (aaRS)-tRNA pairs are typically orthogonal in eukaryotes, our work provides a general strategy to develop additional aaRS-tRNA pairs that can be used for UAA mutagenesis of proteins expressed in both E. coli and eukaryotes.