Machine learning guided batched design of a bacterial Ribosome Binding Site

Optimisation of gene expression levels is an essential part of the organism design process. Fine control of this process can be achieved through engineering transcription and translation control elements, including the ribosome binding site (RBS). Unfortunately, design of specific genetic parts can still be challenging due to lack of reliable design methods. To address this problem, we have created a machine learning guided Design-Build-Test-Learn (DBTL) cycle for the experimental design of bacterial RBSs to show how small genetic parts can be reliably designed using relatively small, high-quality data sets. We used Gaussian Process Regression for the Learn phase of cycle and the Upper Confidence Bound multi-armed bandit algorithm for the Design of genetic variants to be tested in vivo. We have integrated these machine learning algorithms with laboratory automation and high-throughput processes for reliable data generation. Notably, by Testing a total of 450 RBS variants in four DBTL cycles, we experimentally validated RBSs with high translation initiation rates equalling or exceeding our benchmark RBS by up to 34%. Overall, our results show that machine learning is a powerful tool for designing RBSs, and they pave the way towards more complicated genetic devices.

Robust T cell activation requires an eIF3-driven burst in T cell receptor translation

Activation of T cells requires a rapid surge in cellular protein synthesis. However, the role of translation initiation in the early induction of specific genes remains unclear. Here we show human translation initiation factor eIF3 interacts with select immune system related mRNAs including those encoding the T cell receptor (TCR) subunits TCRA and TCRB. Binding of eIF3 to the TCRA and TCRB mRNA 3'-untranslated regions (3'-UTRs) depends on CD28 coreceptor signaling and regulates a burst in TCR translation required for robust T cell activation. Use of the TCRA or TCRB 3'-UTRs to control expression of an anti-CD19 chimeric antigen receptor (CAR) improves the ability of CAR-T cells to kill tumor cells in vitro. These results identify a new mechanism of eIF3-mediated translation control that can aid T cell engineering for immunotherapy applications.

NAP1L5 Promotes Nucleolar Hypertrophy and Is Required for Translation Activation During Cardiomyocyte Hypertrophy

Pathological growth of cardiomyocytes during hypertrophy is characterized by excess protein synthesis; however, the regulatory mechanism remains largely unknown. Using a neonatal rat ventricular myocytes (NRVMs) model, here we find that the expression of nucleosome assembly protein 1 like 5 (Nap1l5) is upregulated in phenylephrine (PE)-induced hypertrophy. Knockdown of Nap1l5 expression by siRNA significantly blocks cell size enlargement and pathological gene induction after PE treatment. In contrast, Adenovirus-mediated Nap1l5 overexpression significantly aggravates the pro-hypertrophic effects of PE on NRVMs. RNA-seq analysis reveals that Nap1l5 knockdown reverses the pro-hypertrophic transcriptome reprogramming after PE treatment. Whereas, immune response is dominantly enriched in the upregulated genes, oxidative phosphorylation, cardiac muscle contraction and ribosome-related pathways are remarkably enriched in the down-regulated genes. Although Nap1l5-mediated gene regulation is correlated with PRC2 and PRC1, Nap1l5 does not directly alter the levels of global histone methylations at K4, K9, K27 or K36. However, puromycin incorporation assay shows that Nap1l5 is both necessary and sufficient to promote protein synthesis in cardiomyocyte hypertrophy. This is attributable to a direct regulation of nucleolus hypertrophy and subsequent ribosome assembly. Our findings demonstrate a previously unrecognized role of Nap1l5 in translation control during cardiac hypertrophy.

Metabolic Plasticity in Melanoma Progression and Response to Oncogene Targeted Therapies

Resistance to therapy continues to be a barrier to curative treatments in melanoma. Recent insights from the clinic and experimental settings have highlighted a range of non-genetic adaptive mechanisms that contribute to therapy resistance and disease relapse, including transcriptional, post-transcriptional and metabolic reprogramming. A growing body of evidence highlights the inherent plasticity of melanoma metabolism, evidenced by reversible metabolome alterations and flexibility in fuel usage that occur during metastasis and response to anti-cancer therapies. Here, we discuss how the inherent metabolic plasticity of melanoma cells facilitates both disease progression and acquisition of anti-cancer therapy resistance. In particular, we discuss in detail the different metabolic changes that occur during the three major phases of the targeted therapy response—the early response, drug tolerance and acquired resistance. We also discuss how non-genetic programs, including transcription and translation, control this process. The prevalence and diverse array of these non-genetic resistance mechanisms poses a new challenge to the field that requires innovative strategies to monitor and counteract these adaptive processes in the quest to prevent therapy resistance.

LARP6 proteins in plants

RNA binding proteins, through control of mRNA fate and expression, are key players of organism development. The LARP family of RBPs sharing the La motif, are largely present in eukaryotes. They classify into five subfamilies which members acquired specific additional domains, including the RRM1 moiety which teams up with the La motif to form a versatile RNA binding unit. The LARP6 subfamily has had a peculiar history during plant evolution. While containing a single LARP6 in algae and non-vascular plants, they expanded and neofunctionalized into three subclusters in vascular plants. Studies from Arabidopsis thaliana, support that they acquired specific RNA binding properties and physiological roles. In particular LARP6C participates, through spatiotemporal control of translation, to male fertilization, a role seemingly conserved in maize. Interestingly, human LARP6 also acts in translation control and mRNA transport and similarly to LARP6C which is required for pollen tube guided elongation, is necessary to cell migration, through protrusion extension. This opens the possibility that some cellular and molecular functions of LARP6 were retained across eukaryote evolution. With their peculiar evolutionary history, plants provide a unique opportunity to uncover how La-module RNA binding properties evolved and identify species specific and basal roles of the LARP6 function. Deciphering of how LARP6, in particular LARP6C, acts at the molecular level, will foster novel knowledge on translation regulation and dynamics in changing cellular contexts. Considering the seemingly conserved function of LARP6C in male reproduction, it should fuel studies aimed at deriving crop species with improved seed yields.

Dysregulated Protein Translation Control in Hidradenitis Suppurativa: Implication for Lesion-associated Squamous Cell Carcinoma Development

Hidradenitis Suppurativa (HS) is a proinflammatory painful skin disorder. This chronic disease is often associated with aggressive squamous cell carcinoma (SCC). The molecular pathogenesis of this disease progression remains undefined. The translation initiation factor eIF4E/4G/4A1 complex is overexpressed in a variety of human malignancies. In this study, we found that the expression of eIF4E/4G/4A1 as well as phosphorylated eIF4E were upregulated in HS skin. In the global transcription profiles derived from two public database, we were able to enrich 734 eIF4F-related genes. GSEA pathway enrichment analysis further demonstrated that RAS/MEK/ERK oncogene signaling pathway associated with inflammation signaling were significantly activated in HS lesion. The increase expression of eIF4 protein components was associated with enhanced eIF4E translation targets Cyclin D1 and c-Myc. Confocal fluorescence microscopy analysis further revealed that Cyclin D1 and c-Myc specifically co-localized in nuclei of certain cells in HS epithelium. We also found that many of the PCNA positive hyperproliferative cells were also positive for c-Myc expression. These data demonstrate that 5-cap-dependent translation is a potential pathway underlying the SCC pathogenesis in chronic HS lesions. Furthermore, being a druggable target, inhibition of eIF4F may block lesion-associated lethal SCCs in HS patients.