Results

Overview

Successful Construction of a Copper Biosensor with Single Plasmid System

We built a plasmid with only the reporter system Fig.1a, which was transferred into E. coli. This system utilizes E. coli's own CueR protein to sense copper ions and output a fluorescent signal from mVenusNB. In order to better simulate the possible environment in a realistic test, we incubated at room temperature of 25 degrees. Here we show the kinetic curves at different copper ion concentrations (Fig. 2a), and we can see that the system exhibits a significant difference from the blank group even at a low copper ion concentration of 170nM within 10hrs. We selected the data points at 35hr for fitting the induced response curve, and it can be found that the reporter system is basically saturated at 250uM Cu2+ concentration(Fig. 2b). This work demonstrate that our design to construct a copper biosensor using endogenous proteins is feasible, revealing the effects of the CueR of chassis bacteria overlooked by previous work utilizing this promoter.

Successful Construction and a Copper Biosensor with Dual Plasmid System
Higher CueR Expression Raises induced change and lowers the leak expression

We then introduced another plasmid for high expression of CueR and carved out the response of this system to different Cu concentrations under the same conditions. Comparing to the above system response curves(Fig. 3a), it can be seen that the system with high expression of CueR still has a better resolution for copper between 250uM-2mM, allowing the system to be used in environments with higher concentrations of copper ions. The system also has lower background leakage(3-fold lower than section I circuit) and higher induced FP multiplicity(Fig. 3b, 4), making it less prone to false positives. Overall, the high expression CueR improves the original copper biosensor

Increased CueR expression lowers the max induced Fluorescence

It is also important to note that increasing CueR expression in our system was found to simultaneously decrease maximal fluorescence expression(Fig. 4), contrary to expectations. We post a potential explanation for this is that since the cell burden is limited, the presence of a highly expressed CueR plasmid forced the intracellular anabolic flow to be split between two exogenous proteins (mVenusNB and CueR) at the same time, which affects the maximal expression of mVenus.

In a finite system like a bacterial cell, the production of exogenous proteins is constrained by the limited availability of cellular resources, such as energy, ribosomes, and amino acids. When multiple proteins are expressed simultaneously, such as CueR and mVenusNB in our system, they compete for these resources. This competition creates a bottleneck in the intracellular anabolic flow, limiting the amount of each protein that can be produced. As a result, even though we would expect increasing CueR expression to enhance activation (and subsequently increase fluorescence), the overall resource limitation actually reduces the maximal expression of mVenus, which explains the drop in fluorescence.

Rational modeling often assumes that higher expression of activators like CueR will lead to a linear or predictable increase in downstream outputs. However, this assumption fails in the context of resource-limited systems, where non-linear effects such as metabolic burden and competition between exogenous proteins can dominate. This is why rational models may fail to predict actual system behavior under such conditions.

To address these complexities, high-throughput screening and machine learning (ML) techniques can be employed. High-throughput screening allows for the rapid testing of multiple combinations of conditions to capture the emergent behaviors that rational models may overlook. ML can then analyze these large datasets to identify patterns and relationships between variables, providing more accurate predictions of how resource competition affects system performance.

Fine-tuning the CueR expression contributes to high-coverage copper receptor libraries

We randomly inserted Promoter and RBS libraries into the CueR 5' end to create a plasmid library with different CueR expression intensities. After cotransforming this plasmid library with the reporter plasmid, we obtained a Cu(II) biosensor library.

We picked 96 strains and measured the kinetic of OD600 and fluorescence at 0uM and 500uM, respectively. In this library, the fluorescence multiplicity before and after induction at 500uM ranged from 5- to 10-fold(Fig.5), showing a broad range of the regulation. 8 of 96 strains had a higher maximum fluorescence after induction than the two systems in sections I & II (>35,000) under the same parameter conditions(Fig. 6). Our project demonstrates the feasibility of tuning the CueR expression to parameterize the circuit and provides a biosensor library for Cu(II) ions.

We observed that all promoters in the promoter library exhibit a certain level of promoter activity, indicating their significant role in transcriptional regulation. The initiation strength of the promoters shows a degree of variability, with the lowest induction strength being 5-fold, and the highest reaching up to 10-fold. To gain a deeper understanding of the relationship between the activity of these promoters and their sequence characteristics, we employed Sanger sequencing to determine the nucleotide sequences of these promoters. Subsequently, based on the sequencing results, we conducted a correlation analysis between the initiation strength of the promoters and their sequence features, and built a mathematical model accordingly. The detailed construction process and parameter settings of this model can be consulted on our provided "Model" page. Through this model, we aim to reveal the specific mechanisms by which promoter sequence characteristics affect their initiation activity, providing a theoretical foundation and experimental guidance for subsequent research on gene expression regulation.

Characterization of the impact of different metal ions on the CueR-pCoA system

We assessed the impact of metal ions (Ca²⁺, Na⁺, K⁺) on the expression of mVenus NB and binding of CueR in our E. coli system. A series of test conditions on 96 plate were prepared and bacterial cultures were exposed to the metal ion solutions for a standardized period. After treating the bacteria with metal ion solutions, we measured the fluorescence intensity of mVenus NB in the bacterial cells for each metal ion treatment. We found that CueR is highly specific to Cu2+ only.

Understanding the effect of Cu2+ concentration on E.coli cell growth

We determined the impact of different Cu2+ concentrations on E.coli growth. E.coli containing the CueR-pCoA system and E.coli with empty plasmid as control were cultured to exponential growth stage and exposed to a series of Cu²⁺ solutions of different concentrations. The florescent intensity curve over time was measured using microplate reader. We found that concentration that <=12.5mM hinders the growth of bacteria. We can take this possibility into account when working with data collected in the field.

The development of a cell-free transcription test strip

We first tested the cell lysis conditions using 75W 15min ultrasonic treatment for 2ml overnight culture. After plating the lysate, we observed that some bacterial cells were not fully lysed. In the second attempt, we increased the power to 90W, which successfully resulted in complete cell lysis as expected.

Next, we added Cu²⁺ to the lysate to test the system's response. However, a white precipitate formed in the solution, which we suspect was caused by the accidental addition of EDTA, leading to a reaction with Cu²⁺.

Due to time constraints and limited experimental conditions, we were unable to complete the functional validation of the system in the lysate. However, we have already designed the hardware for the entire system, aimed at future home-based detection applications. This setup will allow for easy and safe use outside of laboratory conditions, making it suitable for everyday use. The hardware is designed to integrate the cell-free system onto a user-friendly test strip, providing a reliable platform for real-time, at-home biosensing.