Using protein geometry to optimize cytotoxicity and the cytokine window of a ROR1 specific T cell engager.

T cell engaging bispecific antibodies have shown clinical proof of concept for hematologic malignancies. Still, cytokine release syndrome, neurotoxicity, and on-target-off-tumor toxicity, especially in the solid tumor setting, represent major obstacles. Second generation TCEs have been described that decouple cytotoxicity from cytokine release by reducing the apparent binding affinity for CD3 and/or the TAA but the results of such engineering have generally led only to reduced maximum induction of cytokine release and often at the expense of maximum cytotoxicity. Using ROR1 as our model TAA and highly modular camelid nanobodies, we describe the engineering of a next generation decoupled TCE that incorporates a "cytokine window" defined as a dose range in which maximal killing is reached but cytokine release may be modulated from very low for safety to nearly that induced by first generation TCEs. This latter attribute supports pro-inflammatory anti-tumor activity including bystander killing and can potentially be used by clinicians to safely titrate patient dose to that which mediates maximum efficacy that is postulated as greater than that possible using standard second generation approaches. We used a combined method of optimizing TCE mediated synaptic distance and apparent affinity tuning of the TAA binding arms to generate a relatively long but persistent synapse that supports a wide cytokine window, potent killing and a reduced propensity towards immune exhaustion. Importantly, this next generation TCE induced significant tumor growth inhibition in vivo but unlike a first-generation non-decoupled benchmark TCE that induced lethal CRS, no signs of adverse events were observed.

RF1 knockout allows ribosomal incorporation of unnatural amino acids at multiple sites.

Stop codons have been exploited for genetic incorporation of unnatural amino acids (Uaas) in live cells, but their low incorporation efficiency, which is possibly due to competition from release factors, limits the power and scope of this technology. Here we show that the reportedly essential release factor 1 (RF1) can be knocked out from Escherichia coli by 'fixing' release factor 2 (RF2). The resultant strain JX33 is stable and independent, and it allows UAG to be reassigned from a stop signal to an amino acid when a UAG-decoding tRNA-synthetase pair is introduced. Uaas were efficiently incorporated at multiple UAG sites in the same gene without translational termination in JX33. We also found that amino acid incorporation at endogenous UAG codons is dependent on RF1 and mRNA context, which explains why E. coli tolerates apparent global suppression of UAG. JX33 affords a unique autonomous host for synthesizing and evolving new protein functions by enabling Uaa incorporation at multiple sites.

Stereochemical basis for engineered pyrrolysyl-tRNA synthetase and the efficient in vivo incorporation of structurally divergent non-native amino acids.

Unnatural amino acids (Uaas) can be translationally incorporated into proteins in vivo using evolved tRNA/aminoacyl-tRNA synthetase (RS) pairs, affording chemistries inaccessible when restricted to the 20 natural amino acids. To date, most evolved RSs aminoacylate Uaas chemically similar to the native substrate of the wild-type RS; these conservative changes limit the scope of Uaa applications. Here, we adapt Methanosarcina mazei PylRS to charge a noticeably disparate Uaa, O-methyl-l-tyrosine (Ome). In addition, the 1.75 Å X-ray crystal structure of the evolved PylRS complexed with Ome and a non-hydrolyzable ATP analogue reveals the stereochemical determinants for substrate selection. Catalytically synergistic active site mutations remodel the substrate-binding cavity, providing a shortened but wider active site. In particular, mutation of Asn346, a residue critical for specific selection and turnover of the Pyl chemical core, accommodates different side chains while the central role of Asn346 in aminoacylation is rescued through compensatory hydrogen bonding provided by A302T. This multifaceted analysis provides a new starting point for engineering PylRS to aminoacylate a significantly more diverse selection of Uaas than previously anticipated.

Improving orthogonal tRNA-synthetase recognition for efficient unnatural amino acid incorporation and application in mammalian cells.

Optimizing the anticodon recognition between orthogonal tRNA and synthetase significantly increased the incorporation efficiencies of various unnatural amino acids in mammalian cells, and the enhanced incorporation enabled efficient photocrosslinking of interacting proteins in mammalian cells.

Genetically encoding unnatural amino acids for cellular and neuronal studies.

Proteins participate in various biological processes and can be harnessed to probe and control biological events selectively and reproducibly, but the genetic code limits the building block to 20 common amino acids for protein manipulation in living cells. The genetic encoding of unnatural amino acids will remove this restriction and enable new chemical and physical properties to be precisely introduced into proteins. Here we present new strategies for generating orthogonal tRNA-synthetase pairs, which made possible the genetic encoding of diverse unnatural amino acids in different mammalian cells and primary neurons. Using this new methodology, we incorporated unnatural amino acids with extended side chains into the K+ channel Kv1.4, and found that the bulkiness of residues in the inactivation peptide is essential for fast channel inactivation, a finding that had not been possible using conventional mutagenesis. This technique will stimulate and facilitate new molecular studies using tailored unnatural amino acids for cell biology and neurobiology.