Arenavirus RNA genomes are initiated by a "prime and realign" mechanism, such that the initiating GTP is found as a single unpaired (overhanging) nucleotide when the complementary genome ends anneal to form double-stranded (ds) RNA panhandle structures. dsRNAs modeled on these structures do not induce interferon (IFN), as opposed to bluntended 5 ppp dsRNA. This study examines whether these viral structures can also act as decoys, by trapping RIG-I in inactive dsRNA complexes. We examined the ability of various dsRNAs to activate the RIG-I ATPase (presumably a measure of helicase translocation on dsRNA) relative to their ability to induce IFN. We found that there is no simple relationship between these two properties, as if RIG-I can translocate on short dsRNAs without inducing IFN. Moreover, we found that 5 ppp dsRNAs with a single unpaired 5 ppp-nucleotide can in fact competitively inhibit the ability of blunt-ended 5 ppp dsRNAs to induce IFN when co-transfected into cells and that this inhibition is strongly dependent on the presence of the 5 ppp. In contrast, 5 ppp dsRNAs with a single unpaired 5 ppp-nucleotide does not inhibit poly(I-C)-induced IFN activation, which is independent of the presence of a 5 ppp group.

No abstract

Although four ADCs have been approved and over sixty others are in development, the majority contain payloads belonging to two classes; tubulin inhibitors and DNA cross-linkers. Challenges in the development of ADCs include managing off-target toxicity and hydrophobicity. Some payload classes (e.g., PBD dimers) are notably hydrophobic leading to problems (e.g., aggregation) during conjugation. Thus, there is interest in developing novel payloads which retain the potency of DNA cross-linkers but have lower hydrophobicity and a wider therapeutic window when part of an ADC. The pyridinobenzodiazepines (PDDs) are a new class of sequence-selective, DNA mono-alkylating ADC payload which contain a polyheterocyclic chain with sufficient span to guide them to specific DNA sequences (e.g., transcription factor binding sites). The lead PDD payload, FGX-2-62, has a different sequence-selectivity profile to other DNA-interactive agents, spanning 8-9 base-pairs compared to 6-7 for a PBD dimer, and DNA footprinting experiments indicate a preference for 5'-XGXWWWWXX-3' sequences (X is any base; W is A/T). Transcription factor array studies have shown that the molecule inhibits DNA-binding of oncogenic transcription factors (e.g., NF-κB and GATA). In in vitro cell line studies, FGX-2-62 has low pM cytotoxicity in a diverse cell line panel, including stem cells, cells from both solid and blood cancers (e.g., 9 pM in HL-60) and MDR-resistant tumours, and arrests the cell cycle at the G0/G1 phase compared to G2-M arrest for PBD dimers. It is compatible with attachment to most linker technologies, and is significantly less hydrophobic than other payload classes. Initial MTD studies were carried out by separately conjugating (with negligible aggregation) FGX-2-62 and the PBD dimer Talirine to a THIOMAB® version of trastuzumab (DAR = 2). In female athymic nude mice, a greater tolerance was observed for the THIOMAB®-(FGX-2-62) ADC compared to the THIOMAB®-PBD dimer (i.e, MTD >8 mg.kg-1 versus 4 mg.kg-1). In an efficacy study, FGX-2-62 was conjugated to a cancer stem cell-targeting IgG1 antibody (Bstrongximab) with DAR 1.9. Initial evaluation afforded IC50 values of 0.67 nM and 0.47 nM in two antigen positive cell-lines, and an MTD of 6 mg.kg-1 in mice. In an antigen-positive embryonal carcinoma stem cell CDX mouse model, complete regression was observed at a dose of 2 mg.kg-1 (Q7Dx3). In a cholangiocarcinoma PDX model, complete tumour regression was observed out to 80 days (when experiment was terminated) at a dose of 5 mg.kg-1 (Q7Dx3), with no observed toxicity. The favourable hydrophobicity profile of the PDDs and ease of conjugation, along with their novel mechanism of action, significant in vitro cytotoxicity, in vivo efficacy and tolerability in MTD studies suggest that they represent a promising new class of ADC payloads. Citation Format: Nicolas Veillard, Paolo Andriollo, Julia Mantaj, Keith R. Fox, K Miraz Rahman, George Procopiou, Francesco Cascio, David B. Corcoran, Ilona Pysz, Patricia A. Cooper, Steven D. Shnyder, Yawen Ju, Edwin Tan, William M. Schopperle, Paul J. Jackson, David E. Thurston. Pyridinobenzodiazepines (PDDs): A new class of sequence-selective DNA mono-alkylating ADC payloads with low hydrophobicity [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 736.

Although five ADCs have been approved and over sixty others are in development, the majority contain payloads belonging to two classes: tubulin inhibitors and DNA-interactive agents. Challenges in the development of ADCs include managing off-target toxicity and hydrophobicity. Some DNA-interactive payload classes [e.g. pyrolobenzodiazepine (PBD] dimers) are notably hydrophobic, leading to problems such as aggregation during conjugation, and systemic toxicities of the resultant ADCs are also beginning to emerge in clinical settings. Thus, there is interest in developing novel payloads which retain the potency of DNA cross-linking agents but have lower hydrophobicity and a wider therapeutic window when part of an ADC. The pyridinobenzodiazepines (PDDs) are a new class of sequence-selective, DNA mono-alkylating ADC payload, which contain a polyheterocyclic chain with sufficient molecular span and DNA base-pair recognition properties to guide them to specific DNA sequences (e.g. transcription factor binding sites). The favourable hydrophobicity profile of the PDDs and ease of conjugation, along with their novel mechanism of action, significant in vitro cytotoxicity and in vivo tolerability and efficacy when in an ADC format, indicate that they represent a promising new class of ADC payloads.