Liver disease is a global healthcare burden requiring early detection and constant monitoring of pathological biomarkers. However, current diagnosis technologies require heavy equipment and well-trained medical professionals. It limits the accessibility of the general public for liver disease diagnosis only to healthcare facilities. Therefore there is an urgent need for the development of field deployable devices (such as glucose meters for diabetes) as robust, user-friendly and cost-effective diagnosis tools.
The ability of sensing and processing myriad signals makes engineered bacteria an ideal candidate for the development of novel biomedical applications, such as bacterial biosensors for disease diagnosis or environmental monitoring. Current advances in synthetic biology have greatly improved the robustness, sensitivity, signal-to-noise ratio, and signal processing abilities of bacterial biosensors. The self-manufacturing nature of bacteria also offers an advantage for the development of cost-effective devices. However, current development of bacterial biosensors mostly rely on iterative searching and studying novel transcription factors and their corresponding regulatory elements which respond to ligands of interest. In order to accelerate the R & D processes of bacterial biosensors, we decided to develop a scalable and modularized synthetic receptor platform to expand the sensing repertoire of engineered bacteria.
The prototype of EMeRALD (Engineered Modularized Receptors Activated via Ligand - induced Dimerization) is based on swapping the pH sensing domain of E.coli one component system, pH sensor CadC, with a camelid single domain antibody, VHH_Caffeine as its new sensing module. The ligand caffeine induced dimerization of VHH_Caffeine can further dimerize the chimeric receptor and become able to activate downstream reporter gene expression. This prototype enables us to control the gene expression in E.coli simply through food additive caffeine, and helps us to learn fundamental knowledge for tuning the sensitivity or signal-to-noise ratio of synthetic receptors.
Recent studies reveal the potential of serum or urine bile salts as pathological biomarkers for liver diseases. Through literature research, I found that enteropathogenic bacteria such as V.Cholerae sense the presence of bile salts as environmental cues to trigger their virulence pathway while entering into the small intestines. However, engineering V.cholerae as a bile salt sensing device for liver disease diagnosis literally encounters all the conundrums for the development of bacterial biosensors: First, it is a pathogen, which brings the biosafety issues. Second, different from novel bacterial hosts like E.coli, there is no well developed synthetic biology toolbox for biosensing engineering in V.cholerae. Third, the virulence pathway in V.cholerae is not triggered only by bile salts, but also well regulated by other signals such as oxygen, pH, or quorum sensing. In addition, the molecular mechanism of the entire complex signaling pathway has not been fully studied yet.
The utilization of EMeRALD platform for rewiring the bile salt sensing modules from enteropathogenic bacteria enables us to directly engineer and repurpose them in the surrogate host E.coli with the support of synthetic biology and protein engineering toolboxes. First we built the synthetic receptors by rewiring the bile salt sensing modules TcpP/TcpH and VtrA/VtrC from V.cholerae and V.parahaemolyticus into EMeRALD platform, respectively. The EMeRALD platform enables us to characterize their bilt salt specificity profiles with fluorescent signals rather than the toxin expression level. The sensor sensitivity is improved through protein engineering and directed evolution, and a functional key residue within the C-terminal loop of TcpP was further identified through next generation sequencing and sequence-function relationship analysis.
Next we developed the colorimetric output version of the bacterial biosensor. We found that the dynamic and operating ranges can be optimized by adjusting the cell density or incubation time. In addition, the one-pot operation procedure of the colorimetric output version of the bacterial biosensor can greatly simplify the system design while further coupling it to a microfluidic device in the future.
Last but not least, we tested our bactosensor in clinical serum samples from 21 liver transplantation patients. Encouragingly, It takes only 5 μL of serum sample for each test, and we have strong visible colorimetric changes detectable by the naked eyes from the samples of the patients who had a high potential of acute cellular rejection (ACR) after liver transplantation (serum bile acid > 37 μM). And the cost of the consumable materials for tests with three replicates is even less than 1 euros. This bacterial bile salt biosensor with low cost and one-pot experimental procedure encourages us to further promote large scale analysis of clinical samples in the future. It can help us to gain further understanding and to overcome the potential factors which might influence the performance of bacterial biosensors in clinical samples. I sincerely hope that this work will promote the development of robust and cost-effective liver diagnosis devices in the near future.
The works described above have demonstrated the potential of EMeRALD platform for studying, engineering and repurposing natural or artificial sensing modules. In addition, we have also engineered EMeRALD based bacterial biosensors which are able to detect environmental pollutants such as copper and nitrate. We can expect that the EMeRALD platform is able to accelerate the R & D process of bacterial biosensors for diverse applications such as medical diagnosis, environmental monitoring, or controlling devices for therapeutic bacteria. In addition, the capability of providing high throughput screening procedure in a biosafety environment enables EMeRALD platform for screening nature or chemical substances libraries for inhibiting pathological biosensing pathways.