Table1_Surveying the Genetic Design Space for Transcription Factor-Based Metabolite Biosensors: Synthetic Gamma-Aminobutyric Acid and Propionate Biosensors in E. coli Nissle 1917.XLSX

Engineered probiotic bacteria have been proposed as a next-generation strategy for noninvasively detecting biomarkers in the gastrointestinal tract and interrogating the gut-brain axis. A major challenge impeding the implementation of this strategy has been the difficulty to engineer the necessary whole-cell biosensors. Creation of transcription factor-based biosensors in a clinically-relevant strain often requires significant tuning of the genetic parts and gene expression to achieve the dynamic range and sensitivity required. Here, we propose an approach to efficiently engineer transcription-factor based metabolite biosensors that uses a design prototyping construct to quickly assay the gene expression design space and identify an optimal genetic design. We demonstrate this approach using the probiotic bacterium Escherichia coli Nissle 1917 (EcN) and two neuroactive gut metabolites: the neurotransmitter gamma-aminobutyric acid (GABA) and the short-chain fatty acid propionate. The EcN propionate sensor, utilizing the PrpR transcriptional activator from E. coli, has a large 59-fold dynamic range and >500-fold increased sensitivity that matches biologically-relevant concentrations. Our EcN GABA biosensor uses the GabR transcriptional repressor from Bacillus subtilis and a synthetic GabR-regulated promoter created in this study. This work reports the first known synthetic microbial whole-cell biosensor for GABA, which has an observed 138-fold activation in EcN at biologically-relevant concentrations. Using this rapid design prototyping approach, we engineer highly functional biosensors for specified in vivo metabolite concentrations that achieve a large dynamic range and high output promoter activity upon activation. This strategy may be broadly useful for accelerating the engineering of metabolite biosensors for living diagnostics and therapeutics.

工程化益生菌已被提出作为一种无创检测胃肠道生物标志物、探究肠-脑轴的下一代策略。阻碍该策略落地的核心挑战之一，在于难以构建所需的全细胞生物传感器（whole-cell biosensor）。在临床相关菌株中构建基于转录因子的生物传感器（transcription factor-based biosensors），往往需要对遗传元件与基因表达进行大量优化，才能达到所需的动态范围与灵敏度。本研究提出一种高效构建基于转录因子的代谢物生物传感器的方法，该方法借助设计原型构建体快速检测基因表达设计空间，从而筛选出最优遗传设计方案。本研究以益生菌大肠杆菌Nissle 1917（Escherichia coli Nissle 1917, EcN）以及两种神经活性肠道代谢物——神经递质γ-氨基丁酸（gamma-aminobutyric acid, GABA）与短链脂肪酸丙酸盐（short-chain fatty acid propionate）为实验模型，验证了该方法的有效性。本研究构建的EcN丙酸盐传感器利用来自大肠杆菌的PrpR转录激活因子（PrpR transcriptional activator），其动态范围可达59倍，灵敏度提升超过500倍，可匹配生理相关浓度区间。本研究开发的EcN GABA传感器则采用来自枯草芽孢杆菌（Bacillus subtilis）的GabR转录阻遏蛋白（GabR transcriptional repressor），以及本研究中构建的人工GabR调控启动子（synthetic GabR-regulated promoter）。本研究首次报道了用于检测GABA的人工合成微生物全细胞生物传感器，该传感器在EcN中于生理相关浓度下可实现138倍的激活倍数。借助该快速设计原型构建方法，我们成功构建了针对特定体内代谢物浓度的高性能生物传感器，该类传感器可实现较宽的动态范围，且在激活后可产生高活性的启动子输出信号。该策略有望广泛应用于加速面向活体诊断与治疗（living diagnostics and therapeutics）的代谢物生物传感器构建进程。