Biology subjects

Chiu, D.

Publications and source records attributed to Chiu, D..

5 recordsLinked to original sources

bioRxiv2023

Rewiring native post-transcriptional carbon regulators to build multi-layered genetic circuits and optimize engineered microbes for bioproduction

As the bioprocessing and synthetic biology spaces rapidly expand, post-transcriptional regulation is emerging as driver for maximizing signal response rate and for minimizing cost per protein within cells. Designing robust post-transcriptional control systems that have precise tunability and can achieve diverse regulatory outcomes are paramount to advance the field. Herein, we develop a new approach for engineered post-transcriptional control in bacteria by rewiring a native regulatory system in Escherichia coli, the Carbon Storage Regulatory (Csr) Network, to create tunable, complex genetic circuits. First, by co-opting native components of the Csr Network to regulate translation of a target mRNA transcript, we establish a Csr-regulated Buffer Gate. Next, by rationally engineering the interactions between our synthetic construct and the native components of the Csr Network, we expand our original design into a genetic toolbox of 12 Buffer Gates that achieve precise tunability across a 10-fold range of target gene expression. Subsequently, to further regulatory capabilities using this approach, we develop a Csr-regulated NOT Gate through engineering a Csr-activated sequence into our synthetic constructs. We then build upon the Csr Buffer and Not Gates to create post-transcriptional dual input Boolean OR, NOR, AND and NAND Logic Gates, as well as a genetic pulse circuit. As a third step, we demonstrate portability of our Csr-regulated Buffer Gates into three industrially relevant bacteria by recapitulating Buffer Gate activity simply by leveraging the conserved homologous Csr Network in each species. Lastly, as a demonstration of downstream application, we apply our system to a proof-of-concept synthetic mevalonate pathway. Using our engineered constructs, we optimize mevalonate production in E. coli resulting in a three-fold increase in production relative to a transcriptionally controlled mevalonate pathway. As a whole, we establish a novel approach to rewire post-transcriptional regulatory networks for complex bacterial computation that can be utilized for efficient bioproduction in engineered microbes.

bioengineering↗

bioRxiv2021

Transcriptional Regulation of Synthetic Polymer Networks

Individual cells direct non-equilibrium processes through coordinated signal transduction and gene expression, allowing for dynamic control over multicellular, system-wide behavior. This behavior extends to remodeling the extracellular polymer matrix that encases biofilms and tissues, where constituent cells dictate spatiotemporal network properties including stiffness, pattern formation, and transport properties. The majority of synthetic polymer networks cannot recreate these phenomena due to their lack of autonomous centralized actuators (i.e., cells). In addition, non-living polymer networks that perform computation are generally restricted to a few inputs (e.g., light, pH, enzymes), limiting the logical complexity available to a single network chemistry. Toward synergizing the advantages of living and synthetic systems, engineered living materials leverage genetic and metabolic programming to establish control over material-wide properties. Here we demonstrate that a bacterial metal respiration mechanism, extracellular electron transfer (EET), can control metal-catalyzed radical cross-linking of polymer networks. Linking metabolic electron flux to a synthetic redox catalyst allows dynamic, tunable, and predictable control over material formation and bulk polymer network mechanics using genetic circuits. By programming key EET genes with transcriptional Boolean logic, we rationally design computational networks that sense-and-respond to multiple inputs in biological contexts. Finally, we capitalize on the wide reactivity of EET and redox catalyses to predictably control another class of living synthetic materials using copper(I) alkyne-azide cycloaddition click chemistry. Our results demonstrate the utility of EET as a bridge for controlling abiotic materials and how the design rules of synthetic biology can be applied to emulate physiological behavior in polymer networks.

synthetic biology↗

bioRxiv2020

Primary platinum resistance and immune exclusion in ovarian carcinomas with high expression of the homologous recombination mediator RAD51

BackgroundHomologous recombination deficiency (HRD) in ovarian cancer confers increased sensitivity to Poly-ADP-Ribose-Polymerase (PARP) inhibitors and platinum. The homologous recombination (HR) mediator RAD51, however, is commonly overexpressed, potentially driving unregulated HR. Due to non-quantitative measurements and heterogeneous cohorts, the clinical relevance of RAD51 expression in ovarian cancer is unclear. MethodsFluorescent immunohistochemistry and multispectral imaging were used to quantitate RAD51 expression, in relation to other markers, in independent cohorts of ovarian carcinomas from British Columbia Cancer (BCC n=284) and the phase III SCOTROC4 trial (n=268). Independent cohorts (TCGA n=566, GSE9891 n=267, GSE26712 n=185 and GSE3149 n=146) were used for mRNA expression and immune infiltration analyses. ResultsRAD51-High tumours had shorter progression-free and overall survival compared to RAD51-Low cases in both BCC and SCOTROC4. The negative prognostic significance of high RAD51 was primarily evident in cases classified "HRD negative" by the Myriad genomic scar assay. Unexpectedly, overexpression of RAD51 in ovarian cancer cell lines did not affect sensitivity to platinum or PARP inhibitors, but modified the expression of immune-regulatory genes. Accordingly, tumours with high RAD51 mRNA showed consistent changes in immunomodulatory transcripts across four independent ovarian cancer cohorts. In-situ multiplexed imaging confirmed that high RAD51 tumours correlated with tumour exclusion of cytotoxic-T-cells, possibly explaining their poor outcomes after chemotherapy. ConclusionsHigh RAD51 expression, in conjunction with a HRD score <42, identifies a subgroup of EOC cases with platinum resistance and an immune excluded tumour microenvironment. Tumours with high RAD51 may require alternate or additional adjuvant therapeutic strategies to overcome immune exclusion. Statement of Translational RelevanceHomologous recombination deficiency (HRD) in ovarian cancer confers increased sensitivity to Poly-ADP-Ribose-Polymerase (PARP) inhibitors and platinum. Currently, methods to identify patients who do poorly on first-line platinum based chemotherapy in ovarian cancer represent an unmet clinical need. In this report we show that RAD51 expression can identify patients with primary platinum resistance in two large cohorts, including a unique phase-III trial of carboplatin as adjuvant monotherapy. The clinical relevance of a positive genomic scar HRD score, a biomarker for assessing HRD status, in predicting benefit to PARP inhibitors has been established following recent phase 3 trials in first line and 2nd line maintenance treatment of ovarian cancer. Our findings suggest that HRD negative ovarian tumours may be further stratified by RAD51 expression status in the context of platinum sensitivity, and may serve to guide future trials of HRD targeted agents.

cancer biology↗

bioRxiv2020

Synthesis of diagnostic quality cancer pathology images

Deep learning-based computer vision methods have recently made remarkable breakthroughs in the analysis and classification of cancer pathology images. However, there has been relatively little investigation of the utility of deep neural networks to synthesize medical images. In this study, we evaluated the efficacy of generative adversarial networks (GANs) to synthesize high resolution pathology images of ten histological types of cancer, including five cancer types from The Cancer Genome Atlas (TCGA) and the five major histological subtypes of ovarian carcinoma. The quality of these images was assessed using a comprehensive survey of board-certified pathologists (n = 9) and pathology trainees (n = 6). Our results show that the real and synthetic images are classified by histotype with comparable accuracies, and the synthetic images are visually indistinguishable from real images. Furthermore, we trained deep convolutional neural networks (CNNs) to diagnose the different cancer types and determined that the synthetic images perform as well as additional real images when used to supplement a small training set. These findings have important applications in proficiency testing of medical practitioners and quality assurance in clinical laboratories. Furthermore, training of computer-aided diagnostic systems can benefit from synthetic images where labeled datasets are limited (e.g., rare cancers). We have created a publicly available website where clinicians and researchers can attempt questions from the image survey at http://gan.aimlab.ca/.

bioinformatics↗

bioRxiv2019

Oncogenic Mutations in Histologically Normal Endometrium: The New Normal?

The presence of somatic driver mutations in endometriosis has previously been believed to represent early events in transformation, however our group and others have described such mutations in roughly one-third of cases of deep infiltrating or iatrogenic endometriosis. These forms of endometriosis rarely progress to malignancy. Recent studies have also shown somatic driver mutations in normal skin, blood, peritoneal washings, and esophageal epithelium. Such findings prompt speculation on whether such mutations exist in the eutopic endometrium - the likely tissue of origin of endometriosis. In the current study we investigated the presence of somatic driver mutations in histologically normal endometrium from women lacking evidence of gynecologic malignancy or endometrial hyperplasia. Twenty-five women who underwent hysterectomies and 85 women who underwent endometrial biopsies were included in this study. Formalin-fixed, paraffin-embedded tissue specimens were analyzed by means of targeted sequencing followed by orthogonal validation with droplet digital PCR. PTEN and ARID1A immunohistochemistry (IHC) was also performed as surrogates for inactivating mutations in the respective genes. Overall, we observed somatic driver-like events in over 50% of histologically normal endometrial samples analyzed, which included hotspot mutations in KRAS, PIK3CA, and FGFR2 as well as PTEN-loss by IHC. Analysis of anterior and posterior samplings collected from women who underwent hysterectomies is consistent with the presence of somatic driver mutations within clonal pockets spread throughout the uterus. The prevalence of such oncogenic mutations also increased with age (OR: 1.05 (95% CI: 1.00 - 1.10), p = 0.035). These findings have implications on our understanding of aging and so-called "normal tissues", thereby necessitating caution in the utilization of mutation-based early detection tools for endometrial or other cancers.

pathology↗