RESUMEN
Short linear motifs (SLiMs) located in disordered regions of multidomain proteins are important for the organization of protein-protein interaction networks. By dynamic association with their binding partners, SLiMs enable assembly of multiprotein complexes, pivotal for the regulation of various aspects of cell biology in higher organisms. Despite their importance, there is a paucity of molecular tools to study SLiMs of endogenous proteins in live cells. LC3 interacting regions (LIRs), being quintessential for orchestrating diverse stages of autophagy, are a prominent example of SLiMs and mediate binding to the ubiquitin-like LC3/GABARAP family of proteins. The role of LIRs ranges from the posttranslational processing of their binding partners at early stages of autophagy to the binding of selective autophagy receptors (SARs) to the autophagosome. In order to generate tools to study LIRs in cells, we engineered high affinity binders of LIR motifs of three archetypical SARs: OPTN, p62, and NDP52. In an array of in vitro and cellular assays, the engineered binders were shown to have greatly improved affinity and specificity when compared with the endogenous LC3/GABARAP family of proteins, thus providing a unique possibility for modulating LIR interactions in living systems. We exploited these novel tools to study the impact of LIR inhibition on the fitness and the responsiveness to cytarabine treatment of THP-1 cells - a model for studying acute myeloid leukemia (AML). Our results demonstrate that inhibition of LIR of a single autophagy receptor is insufficient to sensitize the cells to cytarabine, while simultaneous inhibition of three LIR motifs in three distinct SARs reduces the IC50 of the chemotherapeutic.
RESUMEN
Protein production in Escherichia coli is a fundamental activity for a large fraction of academic, pharmaceutical, and industrial research laboratories. Maximum production is usually sought, as this reduces costs and facilitates downstream purification steps. Frustratingly, many coding sequences are poorly expressed even when they are codon-optimized and expressed from vectors with powerful genetic elements. In this study, we show that poor expression can be caused by certain nucleotide sequences (e.g., cloning scars) at the junction between the vector and the coding sequence. Since these sequences lie between the Shine-Dalgarno sequence and the start codon, they are an integral part of the translation initiation region. To identify the most optimal sequences, we devised a simple and inexpensive PCR-based step that generates sequence variants at the vector-coding sequence junction. These sequence variants modulated expression by up to 1000-fold. FACS-seq analyses indicated that low GC content and relaxed mRNA stability (ΔG) in this region were important, but not the only, determinants for high expression.