Engineering Success

Overview

In E. coli, there is a well-characterized signaling pathway that regulates intracellular manganese levels. From this manganese homeostatic pathway, the 2022 Wright State iGEM team used the pmntP promoter and the yybP-ykoY riboswitch to make a manganese responsive plasmid. In our original design, we used sfGFP as our reporter protein due to its robust ability to quickly and correctly fold into a working fluorescent protein. To boost the performance of our sensor, we also included a plasmid for the IPTG-inducible expression of the manganese responsive transcription factor mntR. (Both are described in the Results Wiki page.) Our original approach was to use whole cells in a culture-based approach to detect manganese in solution. The double plasmid approach, tested in MG1655 E.coli as pSB3K3-pmntP-riboswitch-sfGFP and pTrc-6X-HIS-mntR, was able to detect manganese down to 0.01mM, but required over 24 hours to complete, produced a small change in fluorescence in response to MnCl₂, and required a plate reader for detection of manganese-induced changes in fluorescence.

To address the limitations from our 2022 approach, we decided to pursue the following engineering steps:

1. DBTL Cycle 1: Change to a cell-free system.
2. DBTL cycle 2: Test a T7 promoter-driven manganese biosensor.
3. DBTL Cycle 3: Test a manganese sensor utilizing a destabilized GFP reporter.
4. DBTL Cycle 4: Test a manganese sensor utilizing a NanoLuciferase (NanoLuc) reporter.

1. Avoid limits due to manganese import and export which occur in a whole-cell approach.
2. Reduce assay time. By changing to a cell-free platform, we were able to avoid the overnight culture of E.coli required in our culture-based approach (2022).
3. Improve fieldability by providing robust, visible reporter output.

DBTL Cycle 1

Change to a cell-free system

Design:

In an effort to improve the performance of the two-plasmid manganese biosensor developed by the 2022 Wright State iGEM team, efforts were undertaken to switch to a cell-free system. This double plasmid system was tested using (1) a commercially available cell-free transcription and translation kit from Arbor Biosciences, the myTXTL sigma70 kit, and (2) a homemade cell-free system (a generous gift from our collaborators at the Air Force Research Laboratories).

Build:

Both the pSB3K3-pmntP-riboswitch-sfGFP primary manganese sensor plasmid and the IPTG-inducible pTrc-6X-HIS-mntR plasmids were constructed by the 2022 Wright State iGEM team. The “build” component of this DBTL cycle was more of a “repurposing” of these plasmids to a cell-free assay approach. High purity and concentration preparations of each plasmid were prepared for use in cell-free assays. The protocols used were provided by the manufacturer (Arbor Biosciences, Inc) or obtained from our collaborators at Air Force Research Laboratories.

Test:

Adapting our assay to a cell-free assay format was ultimately successful, and the homemade cell free system worked better for us than the commercial myTXTL kit. Briefly, the pSB3K3-pmntP-riboswitch-sfGFP plasmid produced a small increase in fluorescence in response to 0.01mM manganese in assays using the homemade cell-free system (Results Wiki page, Figure 8). Interestingly, higher doses (e.g. 1.0mM MnCl₂) appeared to result in quenching of fluorescence, as evidenced by a dose-dependent increase in sfGFP protein from 0.1 to 1.0mM MnCl₂ (Results Wiki page, Figure 9).

Learn:

Our data suggested that the sensor performed similarly in whole-cell and cell-free assays, and laid the foundation for further optimization of the sensor. However, current cell-free testing indicated that current sensor design (pmntP-riboswitch-sfGFP) had low fluorescence and potential quenching at higher doses of MnCl₂, thereby limiting its utility as a rapid and fieldable test. Changing the promoter in the sensor plasmid to the T7 promoter was proposed as a potential means of improving sensor performance.

DBTL Cycle 2

Improving manganese biosensor performance by switching to a T7 promoter

Design:

The team decided to construct a sensor in which the pmntP promoter was changed to T7 to facilitate high-level expression in the hope that increased expression would improve both sensor fluorescent yield and sensitivity.

Build:

We replaced our pmntP promoter in the original manganese sensor plasmid (pSB3K3-pmntP-riboswitch-sfGFP) with a T7 promoter to increase protein production. The switch was made by excising the original geneblock from the pSB3K3 plasmid backbone using EcoRI and SpeI restriction enzymes, and inserting a new geneblock by HiFi cloning. The resulting pSB3K3-T7-riboswitch-sfGFP plasmid retained the ribosomal binding site in the 3’ end of the riboswitch (Figure 1). The plasmid was transformed into BL21(DE3) cells for testing so that IPTG could be used to provide the T7 polymerase needed for transcription from the T7 promoter.

Figure 1: T7 promoter-driven biosensor for the detection of Mn²⁺ contamination of drinking water. The modified version of the manganese sensor incorporated a T7 promoter upstream of the the E.coli manganese-binding riboswitch, ribosome binding site (RBS) and sfGFP reporter in a pSB3K3 plasmid backbone for expression in E.coli.

Test:

We ran a series of whole-cell experiments using methods established by our 2022 team, and a series of cell-free experiments to test the functionality of the T7-riboswitch-sfGFP sensor. Experiments included controls to confirm IPTG induction was successful. However, no manganese-induced fluorescence was observed in either assay format (see Results Wiki page, Part 1, figure 8). Additionally, the addition of recombinant T7 polymerase did not result in sfGFP fluorescence or protein, further confirming that the sensor plasmid was not functional.

Learn:

After testing showed that the T7-promoter driven riboswitch-sfGFP plasmid was not functional, the team looked to see if these findings were consistent with published studies of the pmntP-riboswitch function. A prior study published by Dambach et. al. in 2015 showed that the pmntP-riboswitch yielded a 61.7-fold induction in response to manganese in E.coli whole-cell experiments, but only a 7.9-fold induction when the promoter was switched to T7. They showed similar findings in in vitro transcription assays. Our findings are consistent with their observation that the endogenous pmntP promoter is required for riboswitch function [1]. Since the pmntP promoter appeared to be required for optimal riboswitch function, the team turned to testing alternative reporters.

DBTL Cycle 3

Improving manganese biosensor performance by switching to a deGFP reporter

Design:

Our observation that cell-free tests of the pSB3K3-pmntP-riboswitch-sfGFP sensor showed minimal fluorescence yield, but produced sfGFP protein in a dose-dependent manner prompted the team to consider alternative reporters. The positive control plasmid used in the myTXTL kit utilized a deGFP reporter and produced high levels of fluorescence in cell-free assays. To see if issues associated with fluorescence quenching could be overcome by switching to a more robust fluorescent reporter, the team decided to switch from the original sfGFP reporter to deGFP.

Build:

As in DBTL cycle 2, we utilized the original manganese sensor plasmid (pSB3K3-pmntP-riboswitch-sfGFP) as a source of pSB3K3 plasmid backbone for cloning in a new geneblock. The pSB3K3 backbone was isolated using EcoRI and SpeI restriction enzymes, and the new pmntP-riboswitch-deGFP geneblock was inserted by HiFi cloning. The resulting pSB3K3-pmntP-riboswitch-deGFP plasmid retained all of the functional elements of the original functional sfGFP sensor and an in-frame deGFP coding sequence (Figure 2). The plasmid was transformed into MG1655 cells for testing, mirroring the approach used for the original sfGFP sensor.

Figure 2: Biosensor utilizing a deGFP reporter for the detection of Mn²⁺ contamination of drinking water. The modified version of the 2022 manganese sensor incorporated a deGFP reporter placed downstream of the E.coli pmntP promoter, manganese-binding riboswitch and ribosome binding site (RBS) for expression in E.coli.

Test:

We ran a series of whole-cell experiments using methods established by our 2022 team, and a series of cell-free experiments to test the functionality of the pmntP-riboswitch-deGFP sensor. Switching the reporter from the original sfGFP to deGFP did not increase the fluorescence yield of the sensor in response to manganese (see Results Wiki page, Part 3, figures 23 - 24). In contrast to the original sfGFP sensor plasmid, no deGFP protein could be detected in samples treated with manganese in whole-cell or cell-free experiments.

Learn:

Based on these results, the team concluded that utilizing a fluorescent reporter (sfGFP, deGFP or likely any other GFP derivative) placed downstream of the pmntP promoter and manganese-responsive riboswitch was not likely to yield a robust manganese biosensor even with additional optimization. Although no evidence that manganese is capable of inhibiting GFP fluorescence could be found in the literature, the team opted to switch to a non fluorescent reporter for the next DBTL cycle.

DBTL Cycle 4

Improving manganese biosensor performance by switching to a NanoLuc reporter

Design:

The team sought to further optimize the basic pSB3K3-pmntP-riboswitch-sfGFP sensor by switching to a NanoLuc reporter. In doing so, the team hoped to both avoid fluorescence quenching issues implied by testing in DBTL cycles 2 and 3, and further increase the magnitude of the response to manganese (i.e. increase the fold-change versus no manganese control samples).

Build:

As in previous DBTL cycles, we utilized the original manganese sensor plasmid (pSB3K3-pmntP-riboswitch-sfGFP) as a source of pSB3K3 plasmid backbone for cloning in a new geneblock. The pSB3K3 backbone was isolated using EcoRI and SpeI restriction enzymes, and the new pmntP-riboswitch-NanoLuc geneblock was inserted by HiFi cloning. The resulting pSB3K3-pmntP-riboswitch-NanoLuc plasmid retained all of the functional elements of the original functional sfGFP sensor and an in-frame NanoLuc coding sequence (Figure 3). Again, the plasmid was transformed into MG1655 cells for testing, mirroring the approach used for the original sfGFP sensor.

Figure 3: Biosensor utilizing NanoLuc reporter for the detection of Mn²⁺ contamination of drinking water. The modified version of the 2022 manganese sensor incorporated a NanoLuc reporter placed downstream of the E.coli pmntP promoter, manganese-binding riboswitch and ribosome binding site (RBS) for expression in E.coli.

Test:

The NanoLuc sensor (pSB3K3-pmntP-riboswitch-NanoLuc) exhibited a dose-dependent increase in luciferase output in response to manganese in both whole-cell and cell-free assays (see Results Wiki page Part 3, fig 25 - 29). Although the sensitivity of the sensor was similar to the original deGFP version, both capable of detecting down to 0.01mM MnCl₂, the magnitude of the response was significantly higher with the NanoLuc reporter. Further, samples treated with manganese generated sufficient luminescence to be visualized by eye (at 1mM MnCl₂) and by mobile phone imaging (0.01mM MnCl₂, 30 second exposure).

Learn:

Based on the results from both whole-cell and cell-free tests of the pSB3K3-pmntP-riboswitch-NanoLuc sensor, we conclude that the NanoLuciferase reporter is a significant improvement over prior iterations utilizing deGFP and sfGFP reporters. The NanoLuc sensor provides a greater reporter increase (i.e. fold-change) that is detectable without the need for a lab plate reader.

Implementation

With a working sensor in hand, the team has designed and tested a Luminescence Imaging Device (LID) hardware to enable the end user to test water samples at the point of collection. This hardware is described in detail on the Hardware Wiki page. The hardware is designed for use with a mobile phone application which guides the user through image acquisition and performs the required analysis. Together, we feel that the cell-free approach, mobile device and mobile app represent a major step forward towards a fieldable test for manganese contamination in water samples.

Conclusion

The pSB3K3-pmntP-riboswitch-NanoLuc sensor can be utilized in cell-free assays to detect manganese in water down to 0.01mM manganese (0.5ppm), the limit proposed by O’Neal et.al. in 2015 [2]. Further, the optimized sensor generates sufficient luminescence to be detected and roughly quantitated using a mobile phone.