Project Overview
Our 2022 iGEM team developed an E.coli culture-based biosensor for the detection of manganese contamination in drinking water. The sensor was based on the manganese homeostatic pathway present in E.coli (Figure 1).
Figure 1: E. coli manganese homeostatic pathway. Manganese ions are brought into the cell through the mntH importer. Intracellular manganese ions bind to the mntR transcription factor, promoting its binding to the mntP gene promoter (PmntP) and promoting transcription of the riboswitch-mntP mRNA. Intracellular manganese binding to the riboswitch (RS) causes the riboswitch mRNA to open, allowing ribosome binding and translation of the mntP exporter protein. Figure modified from Wang et al 2015 [1].
The 2022 manganese biosensor approach utilized 2 plasmids as described below.
The main sensor plasmid, pSB3K3-pmntP-riboswitch-sfGFP, had an E.coli manganese responsive pmntP promoter and riboswitch placed upstream of a superfolder GFP (sfGFP) reporter (Figure 2A). This design yielded a sensor with manganese-regulated transcription and translation of the sfGFP reporter protein.
The second plasmid, pTrc-6X-HIS-mntR (Figure 2B), provided inducible expression of a 6X-HIS tagged manganese-binding transcription factor mntR. The inducible mntR plasmid was included to test whether increasing mntR levels would increase sfGFP output from the main sensor plasmid in response to manganese.
Figure 2: Dual plasmid biosensor for the detection of Mn2+ contamination of drinking water.(A) The manganese sensor plasmid incorporated the endogenous E.coli manganese responsive pmntP promoter, manganese-binding riboswitch, a ribosome binding site (RBS) and sfGFP reporter in a pSB3K3 plasmid backbone for expression in E.coli. (B) The inducible mntR plasmid was driven by the tac promoter in the pTrc plasmid and included the lac operon (lacO) for IPTG-inducible expression in E.coli.
The 2022 team carried out tests of the primary sensor plasmid and the inducible mntR plasmid in a whole-cell E.coli assay format. They found that the sensor (Figure 2A) fluoresced in response to 0.1mM manganese. When the inducible mntR plasmid (Figure 2B) was added, the team found that leaky, uninduced expression of mntR (i.e. in the absence of IPTG-induction) improved detection down to 0.01mM manganese. Surprisingly, when IPTG was added and excess mntR protein was produced, no sfGFP was produced at any manganese dose tested. The team theorized that increasing mntR levels caused a downregulation of the mntH manganese importer, and thus intracellular manganese levels.
As configured, the 2022 manganese biosensor had the following limitations:
- Long assay time: 24 hours of culture followed by 8hr of growth with manganese.
- Manganese uptake in live E.coli is limited by endogenous transporters.
- Required the use of live bacteria, limiting fieldability.
- Low fluorescent output: ~2-fold increase in fluorescence after 8hr incubation with 1mM MnCl2.
The 2023 Wright State iGEM team sought to improve the manganese biosensor using the following strategies:
- The team sought to adapt the manganese biosensor to a cell-free platform to avoid the long assay times required by the E. coli culture-based approach. This approach was also intended to circumvent limitations in sensitivity arising from manganese transport occurring in live E.coli. This work is described in PART 1 below.
- To address the low fluorescent output of the original sensor in response to manganese, the team explored the use of alternate promoters (see PART 2 below) and reporters (see PART 3 and 4 below).
- To make the sensor more fieldable, the team also designed a 3D-printable and portable device that enables imaging of luminescence from a mobile phone. This device was designed to be used in conjunction with a mobile phone application for determining manganese chloride levels in the water sample(s) being tested. This work is described in the Wiki Implementation and Special Award: Hardware Wiki.
Cell-Free
Part 1: Adapting and optimizing the manganese biosensor to a cell-free platform
Goal
To improve the performance of the pSB3K3-pmntP-riboswitch-sfGFP manganese sensor developed by the 2022 Wright State iGEM team, we decided to switch to a cell-free system. This approach will (1) shorten assay time, (2) deliver a more reliable readout of manganese levels by avoiding confounding issues of manganese transport in live E. coli, and (3) improve the fieldability by avoiding the use of live bacteria.
Feasibility of the cell-free approach was evaluated by first testing the original 2022 sensor (Figure 2) using both a commercially available myTXTL kit from Arbor Biosciences and a homemade cell-free system. These initial tests served as a proof of concept and the foundational work required for subsequent sensor optimization.
Approach
We first tested the commercial cell-free myTXTL kit from Arbor Biosciences, specifically the sigma70 myTXTL kit designed for transcription and translation, due to its proven performance with most E.coli promoters (Experiments 1-3).
We also tested our 2022 biosensor using a homemade cell-free system. since biosensor performance is known to vary across different cell-free lysates, commercial and home-made (Experiment 4) [2].
Experiment 1: Testing 2022 sensor in commercial cell free myTXTL kit: We tested the effects of increasing doses of MnCl2 (same concentrations used in whole cell assays by the Wright State 2022 team) on the 2022 manganese biosensor (pSB3K3-pmntP-riboswitch-sfGFP) using the Sigma70 myTXTL cell-free kit. We performed the assay according to manufacturer instructions. A negative control (i.e. “-C”, no DNA template) was included as an assay blank. The kit positive control plasmid, p70a(2)-deGFP, which utilized a destabilized GFP (deGFP) reporter was included to confirm the assay was performed correctly. The assay setup was shown in Table 1. The pSB3K3-pmntP-riboswitch-sfGFP sensor plasmid was included at a final concentration of 10nM, selected based on published literature and a preliminary DNA dose curve test (data not shown) [3]
Table 1: Cell-free test of pSB3K3-mntP-riboswitch-sfGFP manganese sensor in commercial myTXTL kit. Reactions were set up in PCR tubes in 12.5µl volumes as indicated, and then transferred to a 96-well V-bottom plate (split into 2 wells of 6µl each) for measurement of sfGFP fluorescence (Ex. 485nm / Em. 515nm) and turbidity (Abs. 600nm) with readings taken every 10 minutes for 24 hours.
Abbreviations: +C: myTXTL kit p70a(2)-deGFP control, Sensor: pSB3K3-mntP-riboswitch-sfGFP.
Results: As expected, the positive control generated significant deGFP fluorescence, but our pSB3K3-mntP-riboswitch-sfGFP sensor did not produce sfGFP fluorescence in the presence or absence of MnCl2 (Figure 3).
Figure 3: myTXTL cell-free test of the pSB3K3-mntP-riboswitch-sfGFP manganese sensor response to 0.01mM – 1mM MnCl2.All points measured as technical duplicates. Error bars indicate +/- 1 standard deviation.
Next, to determine if MnCl2 induced sfGFP protein (despite no observed fluorescence), we performed a Western Blot for GFP protein (Figure 4). We included cell lysate from a culture of E.coli with our pTrc-6X-HIS-mntR plasmid as a negative control, and a sample of affinity purified HIS-tagged eGFP as a positive control to confirm antibody specificity. The no DNA control sample showed no GFP band (lane 1). The kit positive control plasmid showed a very strong deGFP band which was blown out on the blot indicating overloading (lane 2). The biosensor produced sfGFP protein in a dose-dependent manner over the range of MnCl2 concentrations tested (lane 3-6). Finally, the HIS-mntR (negative control) and HIS-eGFP (GFP positive control) samples (lanes 7-8) yielded the expected banding, confirming antibody specificity.
Figure 4: Western blot of samples from myTXTL cell-free test of pSB3K3-mntP-riboswitch-sfGFP manganese sensor.Samples were run on a Western Blot according to our established method, and probed with a mouse anti-GFP monoclonal antibody for the detection of deGFP and sfGFP.
Conclusions from experiment 1:
- Although MnCl2 did not induce sfGFP fluorescence from our pSB3K3-mntP-riboswitch-sfGFP sensor using the commercial myTXTL cell-free kit (Figure 3), it did induce a dose dependent increase in GFP protein (Figure 4).
- We observed some evaporation of test samples in this run. To address this, we filled wells adjacent to the test wells with water as a means of reducing evaporation in future runs.
- We also learned that our positive controls need to be diluted for Western Blot analysis to prevent blowout.
Experiment 2: Determine if IPTG induces expression of 6X-HIS-mntR in the commercial myTXTL cell-free kit.This test was a critical preliminary step required before testing the double plasmid system (i.e. pSB3K3-pmntP-riboswitch-sfGFP sensor and pTrc-6XHIS-mntR, Figure 2) in Experiment 3.
We set up both negative (no DNA) and myTXTL kit positive control samples as described in Experiment 1 to confirm that the assay was performed correctly.
We tested the pTrc-6X-HIS-mntR plasmid with increasing concentrations of IPTG to determine the optimal concentration of IPTG for inducing mntR.
Table 2: Cell-free test of IPTG induction of pTRC-6X-HIS-mntR in commercial myTXTL cell-free kitReactions were set up in PCR tubes in 12.5µl volumes as indicated, and then transferred to a 96-well V-bottom plate (split into 2 wells of 6 µl each) for measurement of sfGFP fluorescence (Ex. 485nm / Em. 515 nm) generated by the positive control and turbidity (Abs. 600nm) with readings taken every 10 minutes for 24 hours.
Abbreviations: +C: myTXTL kit p70a(2)-deGFP control, mntr: pTrc-6X-HIS-mntR.
Results: The positive control generated significant deGFP fluorescence, as expected (data not shown) confirming the assay was set up properly.
We analyzed mntR levels by Western blot performed with an anti-HIS tag antibody for the detection of 6X-HIS-mntR (Figure 5). The control samples lacking the mntR plasmid (-C and +C, lanes 1-2) showed no mntR band. The uninduced (0mM IPTG, lane 3) control pTrc-6X-HIS-mntR sample sample showed no 6X-HIS-mntR, while those treated with 0.1 to 1 mM IPTG (lanes 4-6) showed a modest induction of 6X-HIS-mntR, thus confirming IPTG-inducible expression of mntR using the myTXTL kit. Cell lysate from an E.coli culture expressing the HIS-mntR protein (an archived sample from our 2022 team) showed a HIS-mntR band (lane 7), while a recombinant, purified HIS-tagged eGFP sample showed no HIS-mntR band (lane 8), but showing a lower band corresponding to the cleaved HIS tag MW. These controls confirmed the specificity of the anti-HIS antibody.
Figure 5: Western blot of samples from myTXTL cell-free test of IPTG induction of pTrc-6X-HIS-mntR plasmid. Samples from the run in Table 2 were run on a Western Blot according to our established method, and probed with a rabbit polyclonal anti-HIS tag antibody for the detection of 6X-HIS-tagged mntR (18kDa) and 6X-HIS tag alone (0.8kDa).
Conclusions from experiment 2:
- IPTG induced the expression of 6X-HIS-mntR using the myTXTL cell-free kit, with similar induction achieved with all IPTG concentrations tested. Accordingly, we chose the middle dose of 0.5mM IPTG for use in Experiment 3.
Experiment 3: Determine if the double plasmid approach improves sfGFP yield compared to the sensor alone in the cell-free myTXTL system. Next, we compared the performance of the sensor plasmid (pSB3K3-pmntP-riboswitch-sFGFP) alone to the double plasmid system (pSB3K3-pmntP-riboswitch-sFGFP and pTRC-6X-HIS-mntR) (Table 3). This being a preliminary test, only critical controls were included. (A full run with all critical controls is shown in Experiment 4.) The sensor was tested with and without 0.1mM MnCl2, a dose shown to induce sfGFP protein expression in Experiment 1. The double plasmid was tested alone (i.e. no IPTG or MnCl2) and with 0.1mM MnCl2 in both uninduced and IPTG induced reactions.
Table 3: Cell-free test of myTXTL run of single and double plasmid systems.Reactions were set up in PCR tubes in 12.5µl volumes as indicated, and then transferred to a 96-well V-bottom plate (split into 2 wells of 6 microliters each) for measurement of sfGFP fluorescence (Ex. 485nm / Em. 515 nm) generated by the positive control and turbidity (Abs. 600nm) with readings taken every 10 minutes for 24 hours. We included no DNA and positive control samples per kit instructions and the single sensor with and without MnCl2 as a reference.
Abbreviations: +C: myTXTL kit p70a(2)-deGFP control, Sensor: pSB3K3-mntP-riboswitch-sfGFP, double: pSB3K3-pmntP-riboswitch-sFGFP and pTRC-6X-HIS-mntR.
Results: The no DNA control sample did not fluoresce, while the kit positive control generated significant deGFP, as expected (Figure 6A). The single sensor plasmid alone showed a modest increase in fluorescence in response to 0.1mM MnCl2 over the 24-hour time-course when compared to the no MnCl2 control (Figure 6B). The double plasmid system, by contrast, failed to show a clear induction of sfGFP in response to MnCl2, with or without IPTG induction (Figure 6C).
Figure 6: myTXTL cell-free test of the pSB3K3-pmntP-riboswitch-sfGFP sensor alone and with pTrc-mntR plasmid.All points measured as technical duplicates. Error bars indicate +/- 1 standard deviation.
The above samples were subjected to Western Blot analysis to determine if GFP protein was produced (Figure 7). The myTXTL kit positive control sample showed robust deGFP protein production, as expected (lane 2). None of the sensor alone (lanes 3-4) or double plasmid samples (lanes 5-7) showed any sfGFP protein.
A positive control affinity purified 6X-HIS-eGFP sample showed the expected GFP band. Further, in samples with the mntR plasmid, leaky mntR protein expression was detected in uninduced samples (lane 5-6), but IPTG induction was effective in increasing mntR levels (lane 7). The cell lysate with known HIS-mnTR expression yielded an upper band corresponding to 6X-HIS-mntR and a lower band assumed to be the cleaved HIS-tag.
Figure 7: Immunoblot of cell-free test samples of the pSB3K3-pmntP-riboswitch-sfGFP sensor alone and with pTrc-mntR plasmid.Samples from Table 3 were run on a Western Blot according to our established method, and probed with a mouse monoclonal anti-GFP antibody or the detection of GFP and rabbit polyclonal anti-HIS tag antibody for the detection of 6X-HIS-mntR and the cleaved 6X-HIS tag alone.
Conclusions from experiment 3:
- The sensor plasmid alone showed a modest increase in fluorescence with 0.1mM MnCl2. The induced and IPTG-induced double plasmid approach did not show the same increase.
- pTRC-mntR showed some leaky expression, and was successfully induced by IPTG using the myTXTL cell-free kit.
- Since sensors tend to perform differently with different cell-free mixes (e.g. commercial and homemade), we felt it was important to test our sensor in a homemade system. We were fortunate to collaborate with Drs. Jorge Chavez and Kathryn Beabout (Air Force Research Laboratories) who generously provided us with their homemade cell-free lysate.
Experiment 4: Determine if the double plasmid approach improves sfGFP yield compared to the sensor alone in a homemade cell-free lysate.Tests using the homemade cell-free lysate were performed as described on the Experiments page. The cell-free “master mix” (Table 4) was prepared by combining BL21(DE3) E.coli cell-free lysate with the associated buffer (“mix”). As in prior runs, the experiment included no DNA and the myTXTL kit positive control (reactions 1 and 2, respectively). A MnCl2 dose-curve was tested for the sensor plasmid alone (reactions 3-6) and in combination with the pTrc-6X-HIS-mntR plasmid (“double plasmid” reactions 9-14). Since the pTrc-6X-HIS-mntR plasmid is IPTG-inducible, we tested both uninduced (reactions 9-11) and IPTG induced (reactions 12-14) conditions. Additionally, we included the pTrc-6X-HIS-mntR plasmid alone without and with IPTG (reactions 7 and 8) to confirm IPTG induction of the HIS-mntR protein.
Table 4 : Cell-free test of dual plasmid system using homemade cell-free kit.Reactions were set up in PCR tubes in 12.5µl volumes as indicated, and then transferred to a 96-well V-bottom plate (split into 2 wells of 6µl each) for measurement of sfGFP fluorescence (Ex. 485nm / Em. 515 nm) generated by the positive control and turbidity (Abs. 600nm) with readings taken every 10 minutes for 24 hours.
Abbreviations: +C: myTXTL kit p70a(2)-deGFP control, Sensor: pSB3K3-mntP-riboswitch-sfGFP, mntR: pTrc-6X-HIS-mntR.
Results: The no DNA control sample did not fluoresce, while the kit positive control generated significant deGFP, as expected (Figure 8A). Testing of the sensor alone, showed that 0.01 mM MnCl2, increased fluorescence compared to the no manganese control sample, 0.1mM had no effect, and 1mM MnCl2 inhibited fluorescence (Figure 8B). For the double plasmid samples, we observed an increase in fluorescence with 0.01mM MnCl2 compared to the 0mM MnCl2 control in the uninduced samples, but induction with 1mM IPTG caused a reduction to the levels observed in the 0mM MnCl2 control (Figure 8C). The double plasmid response to 0.1mM MnCl2 was similar to the 0.01mM MnCl2 response (data not shown). In the high 1mM MnCl2 treated samples, no increase in fluorescence was observed in either uninduced or IPTG induced samples (Figure 8D).
Figure 8: Cell-free test of the pSB3K3-pmntP-riboswitch-sfGFP sensor alone and with pTrc-mntR plasmid using homemade cell-free mix. All points measured as technical duplicates. Error bars indicate +/- 1 standard deviation.
The above samples were subjected to Western Blot analysis to determine if GFP protein was produced (Figure 9). The kit positive control (lane 1, both blots) showed very strong deGFP expression, and the no DNA control did not (lane 2, both blots), as expected.
Samples with the single sensor alone showed no sfGFP protein production in the absence of MnCl2 (lane 3, left blot), and a dose-dependent increase in sfGFP protein in response to MnCl2 (0.1mM and 1.0mM, lanes 5-6, left blot).
The double plasmid samples showed the same dose-dependent increase in sfGFP protein in both uninduced (right blot, lanes 4-5) and with IPTG induction of 6X-HIS-mntR (right blot, lanes 7-8). Leaky expression of the mntR protein was observed in the uninduced double pTrc-6X-HIS-mntR plasmid, both at low levels in the absence of IPTG (right blot lanes 3-5), and at higher levels in the presence of IPTG (right blot lanes 6-8). Whole-cell extract from IPTG-induced E.coli carrying the pTrc-6X-HIS-mntR plasmid showed the expected 6X-HIS-mntR band (lane 9, both blots).
Figure 9: Immunoblot of cell-free test samples using the pSB3K3-pmntP-riboswitch-sfGFP sensor alone and with pTrc-mntR plasmid.Samples from the run in Table 4 were run on a Western Blot according to our established method and probed with mouse monoclonal anti-GFP and rabbit polyclonal anti-HIS tag antibodies for the detection of GFP proteins and HIS-tagged mntR, respectively.
T7 Sensor
Part 2: Testing an inducible T7 promoter-driven manganese sensor (T7-riboswitch-sfGFP)
Goal
The pmntP promoter used in the 2022 sensor (pSB3K3-pmntP-riboswitch-sfGFP, Figure 2A) yielded increased fluorescence with 0.1mM MnCl2, but the modest fluorescence yield required the use of a plate reader for detection, once again limiting the fieldability of the biosensor. The team decided to test a version that utilized a T7 promoter in the hope that increased expression would improve both sensor fluorescent yield and sensitivity.
Approach
The geneblock (pmntP-riboswitch-sFGFP) used to construct the 2022 manganese biosensor was cut from the biosensor plasmid (pSB3K3-pmntP-riboswitch-sFGFP) and replaced with a new T7-riboswitch-sFGFP genblock using EcoR1 and SpeI cut sites to yield the T7 driven biosensor shown in Figure 10.
Figure 10: 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.
Cloning of T7-riboswitch-sfGFP sensor into pSB3K3: The pSB3K3-T7-riboswitch-sfGFP plasmid was prepared by inserting a T7-riboswitch-sfGFP geneblock into the linearized pSB3K3 backbone as follows:
(1) Isolation of the linearized pSB3K3 plasmid backbone: To isolate the pSB3K3 plasmid backbone for HiFi cloning, we digested the 2022 pSB3K3-pmntP-riboswitch-sfGFP sensor plasmid with EcoRI and SpeI-HF (as described in the Experiments Wiki page) to excise the original geneblock. This was followed by gel purification of the linearized pSB3K3 backbone.
Results: The uncut pSB3K3-pmntP-riboswitch-sfGFP plasmid displayed multiple bands corresponding to nicked, linear and supercoiled, as expected (Figure 11). Single enzyme digests exhibited a single band matching the linearized plasmid (4,401bp), thus confirming that each of the enzymes cut effectively. The double digested plasmid DNA yielded the expected pSB3K3 backbone and 2022 sensor geneblock (pmntP-riboswitch-sfGFP) bands as indicated. The 2,649bp linearized pSB3K3 backbone was cut out and purified using the Monarch DNA Gel Extraction Kit.
Figure 11: Agarose gel electrophoresis of restriction digest of pSB3K3-pmntp-riboswitch-sfGFP for isolation of the pSB3K3 backbone. Plasmid DNA was digested with EcoRI and SpeI for 3 hours and then run on an 0.8% low-melting temperature agarose gel at ~100V.
(2)HiFi cloning of the T7-riboswitch-sfGFP geneblock into the linearized pSB3K3 plasmid backbone: We next ran the linearized pSB3K3 plasmid (shown in Figure 11) and resuspended geneblock on an agarose gel to visualize the size and relative concentration of DNA fragments in preparation for HiFi cloning (Figure 12). We then used the gel to estimate a 2:1 ratio of backbone to insert (determined to be 0.5μl of backbone and 5μl of insert) for use in HiFi cloning. We also included a second reaction in which the 2:1 ratio was calculated by nanodrop quantitation.
Figure 12: Agarose gel electrophoresis of linearized pSB3K3 backbone and T7-riboswitch-sfGFP geneblocks. A 0.8% agarose gel was run at ~100V using the sample volumes indicated.
HiFi cloning was performed according to the protocol described on the Experiments Wiki page. A no DNA negative control was included to confirm that antibiotic selection was effective. The kit positive control (consisting of 2 DNA fragments) was included as a confirmation that the kit itself was working. An additional pUC19 positive control was used as a readout of competent cell efficiency.
Results: The kit positive control (consisting of 2 DNA fragments) yielded >100 small colonies (difficult to see in Figure 13), confirming that the kit worked as expected. The lack of colonies on the "no DNA" negative control confirmed that antibiotic selection was effective. The successful transformation of pUC19 confirmed competency of the NEB5α competent cells used. Finally, both HiFi cloning reactions for our pSB3K3-T7-riboswitch-sfGFP target plasmid resulted in lots of colonies for subsequent screening by restriction digestion and sequencing.
Figure 13: Results of transformation of HiFi cloning of pSB3K3-T7-riboswitch-sfGFP. Transformations were performed according to kit instructions utilizing LB plates with Kanamycin or Ampicillin (as appropriate).
Next, we selected five colonies, cultured them, and proceeded to prepare minipreps for screening through restriction digestion (Figure 14). Restriction digests of all five selected colonies yielded bands corresponding to the pSB3K3 backbone and the T7 geneblock insert. As an additional control, we included the 2022 sensor, and that digest also showed the expected bands.
Figure 14: Agarose gel electrophoresis screen of colonies from HiFi cloning run to confirm cloning of the pSB3K3-T7-riboswitch-sfGFP plasmid. Plasmid DNA was digested with EcoRI and SpeI for 3 hours and run on an 0.8% agarose gel at ~100V until sufficient separation of bands was achieved.
We submitted the plasmid for sequencing confirmation. The subsequent sequencing analysis of DNA from colony #1 confirmed the sequence accuracy of the T7-riboswitch-sfGFP insert including the flanking cut sites and plasmid backbone. The pSB3K3-T7-riboswitch-sfGFP plasmid was transformed into BL21(DE3) cells for functional testing in whole-cell and cell-free tests. (BL21(DE3) E.coli were used to enable IPTG induction of T7 polymerase to drive expression from the T7 sensor.)
(3)Functional testing of the pSB3K3-T7-riboswitch-sfGFP sensor plasmid:
Experiment 5: Determine if the pSB3K3-T7-riboswitch-sfGFP sensor improves fluorescence produced in response to manganese compavred to the 2022 sensor in a whole-cell assay. We performed a whole-cell test comparing the new T7 sensor (pSB3K3-T7-riboswitch-sfGFP) (Figure 10) to the 2022 sensor (Figure 2A) using a standard whole-cell assay as described in the Experiments Wiki page. Briefly, we added 1mM MnCl2 to log-phase cultures (0D600 of 0.5) and measured fluorescence (485/515) and OD600 every 2 hours through an 8-hour endpoint (as described in Figure 15A). 1mM IPTG was added to the T7 sensor samples to induce T7 polymerase. This dose was selected based on prior experience from our 2022 team with whole-cell assays showing 1mM IPTG induced the mntR plasmid (Figure 2B) and other inducible sensors. pET28b-6X-HIS-eGFP was included as a control to confirm IPTG induction was effective (shown in the Western blot only).
Results: Cultures of the pSB3K3-pmntP-riboswitch-sfGFP 2022 sensor (in MG1655) exposed to 1mM MnCl2 showed the expected increase in fluorescence relative to no MnCl2 controls over the 8-hour period (Figure 15). The same sensor in BL21 cells also increased in fluorescence with 1mM MnCl2 treatment vs. no MnCl2 controls, but not to the magnitude observed in MG1655. By contrast, the T7 sensor induced with IPTG in BL21(DE3) showed no fluorescence increase, preliminarily suggesting that the T7 sensor was not functional.
Figure 15: Whole cell testing of the new pSB3K3-T7-riboswitch-sfGFP sensor. All measurements made in biological duplicate and read in technical triplicate. Error bars indicate +/- 1 standard deviation.
The above experiment was repeated 2 more times with added control and MnCl2 dose curves, and again no increase in fluorescence was observed from the T7 sensor (data not shown).
A Western blot was performed to determine if GFP protein was produced by the T7 sensor (Figure 16). The T7 sensor did not show a clear increase in GFP protein expression in response to 1mM MnCl2 (lanes 3-4) compared to controls (lanes 1-2). The 2022 sensor, by contrast, showed higher GFP protein expression in cultures treated with 1mM MnCl2 in BL21(DE3) (lane 5) and MG1655 (lane 7) compared to the no MnCl2 control (lane 6). A robust induction of GFP protein was observed in parallel cultures of pET28b-6X-HIS-eGFP expressing E.coli, confirming that IPTG induction was effective, and a robust induction was observed (lanes 8-9).
Figure 16: Immunoblot of whole-cell culture test samples from cultures carrying the new pSB3K3-T7-riboswitch-sfGFP sensor or the 2022 pSB3K3-pmntP-riboswitch-sfGFP sensor. Samples from the run in Figure 15 were run on a Western Blot according to our established method, and probed with a mouse monoclonal anti-GFP antibody for the detection of GFP.
Conclusions from experiment 5:
- The T7 sensor does not appear to be functional in whole-cell assays.
Experiment 6: Determine if the pSB3K3-T7-riboswitch-sfGFP improves fluorescence produced in response to manganese compared to the 2022 sensor in a cell-free assay. A cell-free test of the T7 sensor was performed with and without 1mM MnCl2 treatment (Table 5). Since the cell-free lysate used was from a BL21(DE3) E.coli culture that had not been induced with IPTG to drive T7 polymerase expression, we tested the pSB3K3-T7-riboswitch-sfGFP plasmid with and without recombinant T7 polymerase (rT7) in this run. We also included a pY71-T7-sfGFP plasmid with and without rT7, both as an assay positive control and to confirm that the rT7 addition was effective.
Table 5: Cell-free test of pSB3K3-T7-riboswitch-sfGFP manganese sensor response to 1mM MnCl2.Reactions were set up in PCR tubes in 12.5µl volumes as indicated, and then transferred to a 96-well V-bottom plate (split into 2 wells of 6µl each) for measurement of sfGFP fluorescence (Ex. 485nm / Em. 515 nm) and turbidity (Abs. 600nm) with readings taken every 10 minutes for 24 hours.
Abbreviations: +C: myTXTL kit p70a(2)-deGFP control, T7 sensor: pSB3K3-T7-riboswitch-sfGFP, pY71: T7 control plasmid pY71-T7-sfGFP, rT7: recombinant T7 polymerase.
Results: Due to an error with the plate reader in this run, the fluorescence time-course data was lost. Accordingly, only data from the Western blot is shown here (Figure 17). No GFP protein was produced by the pSB3K3-T7-riboswitch-sfGFP sensor, both in the absence (lanes 1-2) or presence of added rT7 (lanes 3-4). The pY71-T7-sfGFP positive control showed robust expression of sfGFP, both in the absence and presence of rT7 (lanes 5 and 6, respectively).
Figure 17: Immunoblot of samples from the pSB3K3-T7-riboswitch-sfGFP cell-free run.Samples from the run in Table 5 were run on a Western Blot according to our established method and probed with a mouse monoclonal anti-GFP antibody for the detection of deGFP and sfGFP.
Conclusion from Experiment 6
- The T7 sensor does not appear to be functional in cell-free assays.
- Based on the lack of detectable manganese-induced fluorescence or GFP protein in both whole-cell and cell-free tests of the pSB3K3-T7-riboswitch-deGFP sensor plasmid, the team decided to pursue testing of a pmntP-riboswitch geneblock paired with other reporter proteins.
deGFP Sensor
Part 3: Testing a manganese sensor utilizing deGFP (pSB3K3-pmntP- riboswitch-deGFP)
Goal
The low fluorescence yield of the pmntP promoter used in the 2022 sensor (pSB3K3-pmntP-riboswitch-sfGFP, Figure 2A) and the failure of the pSB3K3-T7-riboswitch-sfGFP sensor (Figure 10) to yield fluorescence or protein in response to MnCl2 prompted the team to test other reporter proteins. The deGFP reporter was selected because the myTXTL kit positive control plasmid worked well in cell-free assays suggesting deGFP might not be as prone to quenching or misfolding when placed upstream of the manganese biosensor. Additionally, NanoLuciferase (“NanoLuc”) was selected for testing due to the high sensitivity of luciferase reporter proteins. Both plasmids were cloned in parallel and as described below. The functional testing of the NanoLuc sensor plasmid is described in PART 4.
Approach
The geneblock (pmntP-riboswitch-sFGFP) used to construct the 2022 manganese biosensor was cut from the biosensor plasmid (pSB3K3-pmntP-riboswitch-sFGFP) and replaced with new pmntP-riboswitch-deGFP or pmntP-riboswitch-NanoLuc genblocks using EcoR1 and SpeI cut sites to yield the biosensors shown in Figure 18.
Figure 18. Biosensors utilizing deGFP and NanoLuc reporters for the detection of Mn2+ contamination of drinking water. The modified version of the 2022 manganese sensor incorporated deGFP (A) or NanoLuc (B) reporters downstream of the E.coli pmntP promoter, manganese-binding riboswitch and ribosome binding site (RBS) for expression in E.coli.
Cloning of pmntP-riboswitch-deGFP and NanoLuc sensors into pSB3K3: The pSB3K3-pmntP-riboswitch -deGFP and NanoLuc versions of the manganese biosensor were prepared by inserting geneblocks into the linearized pSB3K3 backbone (mirroring the approach used for the T7 sensor in PART 2) as follows:
(1) Isolation of the linearized pSB3K3 plasmid backbone: To isolate the pSB3K3 plasmid backbone for HiFi cloning, we repeated the strategy described in Figure 11. Briefly, we digested the 2022 pSB3K3-pmntP-riboswitch-sfGFP sensor plasmid with EcoRI and SpeI-HF (as described in the Experiments Wiki page) to excise the original geneblock. This was followed by gel purification of the linearized pSB3K3 backbone.
Results: The 2,649bp linearized pSB3K3 backbone was successfully isolated and purified for HiFi cloning of the T7 sensor and NanoLuc sensor plasmids.
(2) HiFi cloning of the T7-riboswitch-sfGFP geneblock into the linearized pSB3K3 plasmid backbone: We next ran the resuspended geneblock samples and the linearized pSB3K3 plasmid on an agarose gel to visualize the size and relative concentration of DNA fragments for HiFi cloning (Figure 19). We used the gel to estimate a 2:1 ratio of backbone to insert (determined to be 0.5μl of backbone and 5μl of insert) for HiFi cloning.
Figure 19: Agarose gel electrophoresis of linearized pSB3K3 backbone and deGFP and NanoLuc geneblocks.A 0.8% agarose gel was run at 90V using the sample volumes indicated.
HiFi cloning was subsequently performed according to the protocol described on the Experiments Wiki page. No DNA, kit positive control and pUC19 controls were included as described previously.
Results: All controls worked as expected. HiFi cloning reactions for our pSB3K3-pmntP-riboswitch -deGFP and pSB3K3-pmntP-riboswitch-NanoLuc target plasmids resulted in lots of colonies for subsequent screening by restriction digestion and sequencing (Figure 20).
Figure 20: Transformation of HiFi cloning of pSB3K3-pmntP-riboswitch-deGFP and NanoLuc. Transformations were performed according to kit instructions utilizing LB plates with Kanamycin or Ampicillin (as appropriate).
Next, we selected four colonies from each of the newly cloned sensors, cultured them, and proceeded to prepare minipreps for screening through restriction digestion (Figure 21). Restriction digests of all eight colonies yielded bands corresponding to the pSB3K3 backbone and the NanoLuc or deGFP geneblock inserts. As an additional control, we included the 2022 sensor, and that digest also showed the expected bands.
Figure 21: Agarose gel electrophoresis screen of colonies from HiFi cloning run to confirm cloning of both pSB3K3-pmntP-riboswitch-deGFP and NanoLuc sensor plasmids. Plasmid DNA was digested with EcoRI and SpeI for 3 hours and run on an 0.8% agarose gel at ~100V until sufficient separation of bands was achieved.
We submitted the plasmid for sequencing confirmation. The subsequent sequencing analysis confirmed the sequence accuracy of both the pmntP-riboswitch-deGFP and pmntP-riboswitch-NanoLuc plasmids.
Whole-cell and cell-free tests were performed for the pSB3K3-pmntP-riboswitch-deGFP sensor (below) and pSB3K3-pmntP-riboswitch-NanoLuc sensor (Part 4).
Functional testing of the pSB3K3-pmntP-riboswitch-deGFP sensor
Experiment 7: Determine if the pSB3K3-pmntP-riboswitch-deGFP sensor improves fluorescence produced in response to manganese compared to the 2022 sensor in a whole-cell assay. We performed a whole-cell test comparing the new deGFP sensor (pSB3K3-pmntP-riboswitch-deGFP) (Figure 18A) to the 2022 sensor (Figure 2A) using a standard whole-cell assay as described in the Experiments Wiki page.
Briefly, overnight cultures of MG1655 E.coli carrying the target sensor plasmids were diluted 1:20 in LB media with Kanamycin selection, allowed to grow to an OD600 of 0.5, and then treated with the 0, 0.1 or 1.0mM MnCl2. A positive control sample of pET29b(+)-6X-HIS-eGFP was included with and without 1mM IPTG treatment. Fluorescence (485nm Excitation / 515nm Emission) and absorbance (A600) measurements were taken every 2 hours until the 6 hour time point. All measurements were collected in technical triplicate.
Results: The uninduced positive control did not fluoresce, while the IPTG-induced positive control generated significant deGFP, as expected (Figure 22A).The 2022 sfGFP sensor (assay control) showed a dose-dependent increase in fluorescence in response to 0.01mM and 1mM MnCl2 compared to 0mM MnCl2 controls evident by 2 hours and persisting through 6hr (Figure 22B). Cultures of the new deGFP sensor showed no MnCl2 induced fluorescence (Figure 22C).
Figure 22: Whole-cell testing of the new pSB3K3-pmntP-riboswitch-deGFP vs. the original pSB3K3-pmntP-riboswitch-sfGFP sensor response to MnCl2The fold-change increase in fluorescence for the (A) IPTG induced pET29b(+)-6X-HIS-eGFP control relative to an uninduced control, and 0.1mM or 1mM MnCl2 treated (B) pSB3K3-pmntP-riboswitch-sfGFP and (C) pSB3K3-pmntP-riboswitch-deGFP relative to untreated controls. All measurements made in technical triplicate. Error bars indicate +/- 1 standard deviation.
A Western Blot of the samples in Figure 23 showed that no deGFP protein was evident in the untreated or any of the MnCl2 treated samples with the pSB3K3-pmntP-riboswitch-deGFP sensor (lanes 1-4) (Figure 23). Cultures with the original pSB3K3-pmntP-riboswitch-sfGFP sensor showed no sfGFP expression in the untreated or low doses of MnCl2, but showed a clear induction of sfGFP with 1.0mM MnCl2, (lanes 5-8), generally consistent with prior runs. The IPTG-induced positive control sample (lane 9), showed strong GFP expression, as expected.
Figure 23: Immunoblot of whole-cell culture test samples from cultures carrying the new pSB3K3-pmntP-riboswitch-deGFP vs the 2022 pSB3K3-pmntP-riboswitch-sfGFP sensor. A Western Blot was run according to our established method, and probed with a mouse monoclonal anti-GFP antibody for the detection of deGFP, sfGFP and eGFP.
Conclusion from Experiment 7:
- The deGFP sensor does not appear to be functional in whole-cell assays.
Experiment 8: Determine if the pSB3K3-pmntP-riboswitch-deGFP improves fluorescence produced in response to manganese compared to the 2022 sensor in a cell-free assay. A cell-free test of the pSB3K3-pmntP-riboswitch-deGFP sensor compared to the 2022 pSB3K3-pmntP-riboswitch-deGFP sensor was performed (Table 6). As in prior runs, we set up both negative (no DNA) and myTXTL kit positive control samples to confirm that the assay was performed correctly. We tested both plasmids with and without 1mM MnCl2 treatment.
Table 6: Cell-free test of pSB3K3-pmntP-riboswitch-deGFP and pSB3K3-mntP-riboswitch-sfGFP manganese sensor response to MnCl2.Reactions were set up in PCR tubes in 12.5µl volumes as indicated, and then transferred to a 96-well V-bottom plate (split into 2 wells of 6µl each) for measurement of sfGFP fluorescence (Ex. 485nm / Em. 515 nm) and turbidity (Abs. 600nm) with readings taken every 10 minutes for 24 hours.
Abbreviations: +C: myTXTL kit p70a(2)-deGFP control, sfGFP Sensor: pSB3K3-mntP-riboswitch-sfGFP, deGFP Sensor: pSB3K3-mntP-riboswitch-deGFP
Results: The deGFP did not produce fluorescence in response to 1mM MnCl2 in the cell-free assay (data not shown). The corresponding western blot (Figure 24) showed the expected sfGFP protein produced by the myTXTL kit positive control plasmid (lane 2) and pSB3K3-pmntP-riboswitch-sfGFP sensor with 1mM MnCl2 (lane 5). However, no GFP protein was produced by the pSB3K3-pmntP-riboswitch-deGFP sensor, either with or without 1mM MnCl2 added (lanes 7-8).
Figure 24: Immunoblot of cell-free test samples from cultures carrying the new pSB3K3-pmntP-riboswitch-deGFP vs. the 2022 pSB3K3-pmntP-riboswitch-sfGFP sensor. A Western Blot was run according to our established method, and probed with a mouse monoclonal anti-GFP antibody for the detection of sfGFP and deGFP.
Conclusion from Experiment 8:
- The deGFP sensor does not appear to be functional in cell-free assays.
- Based on the lack of detectable manganese-induced fluorescence or GFP protein in both whole-cell and cell-free tests of the pSB3K3-pmntP-riboswitch-deGFP sensor plasmid, the team abandoned further testing of the deGFP sensor and began testing of the NanoLuc sensor.
NanoLuc Sensor
Part 4: Functional testing of the pSB3K3-pmntP-riboswitch-NanoLuc sensor
The cloning of the pSB3K3-pmntP-riboswitch-NanoLuc sensor plasmid is described in PART 3 above.
Experiment 9: Determine if the pSB3K3-pmntP-riboswitch-Nanoluc sensor responds to MnCl2 in a dose-dependent manner in a whole-cell assay. The response of the NanoLuc sensor (pSB3K3-pmntP-riboswitch-Nanoluc, Figure 18B) to a 6-hour treatment with 0mM, 0.01mM, 0.1mM, 1mM and 10mM MnCl2 was determined. The assay was set up as described in the Experiments Wiki page. An inducible NanoLuc plasmid (pET-28a(+)::NL) was included to serve as a positive control.
The NanoLuc sensor response to MnCl2 was tested using 10µl aliquots of the culture with 10µl of furimazine substrate. The positive control plasmid yielded strong luminescence (data not shown). The NanoLuc sensor showed a dose-dependent response to MnCl2 (Figure 25). The lowest dose tested, 0.01mM MnCl2, yielded a >2.4 fold increase in luminescence relative to the 0mM MnCl2 control. All test samples yielded measurable luminescence levels within the linear range of the BioTek Synergy H1 plate reader that were stable for at least 15 minutes (data not shown).
Figure 25: Whole-cell testing of the pSB3K3-pmntP-riboswitch-NanoLuc sensor response to MnCl2 .Dose-response test NanoLuc sensor to MnCl2. All measurements made without replication (i.e. screening test only).
Conclusions from Experiment 9:
- The NanoLuc sensor showed a dose-dependent response to MnCl2 from 0.01mM - 10mM MnCl2 in the whole-cell assay.
Experiment 10: Determine if the pSB3K3-pmntP-riboswitch-Nanoluc sensor responds to MnCl2 in a dose-dependent manner in a cell-free assay format.Next, we tested the response of the Nanoluc sensor to MnCl2 in a cell-free assay format (Table 7). A no plasmid DNA sample was included as a negative control, and the inducible NanoLuc plasmid (pET-28a(+)::NL) served as a positive control (+C). Luminescence measurements were taken at 2hr, 4hr, 6hr and 24hr to determine the assay time required for sufficient luminescence detection.
Table 7: Cell-free test of pSB3K3-pmntP-riboswitch-NanoLuc manganese sensor response to MnCl2.Reactions were set up in PCR tubes in 12.5µl volumes as indicated and incubated at 29C. Luciferase measurements were performed with the NanoGlo Luciferase Assay kit using 1µl sample + 9µl water + 10µl furimazine substrate in a white plate.
Abbreviations: -C: no plasmid DA (negative) control; Nanoluc +C: pET28a(+)::NL, NanoLuc Sensor: pSB3K3-mntP-riboswitch-NanoLuc.
Results: Significant luminescence was observed in the positive control sample (data not shown). The Nanoluc sensor samples showed a stable luminescence level 20 minutes beyond substrate addition (data not shown). A consistent fold-change in luminescence from the 0mM MnCl2 control was observed across all timepoints (Figure 26A), indicating a 2 hour assay was feasible. Experiment 10 was repeated 2 additional times, and the combined data showed a dose-dependent increase in luminescence from 0.01mM - 1mM (Figure 26B).
Figure 26: Cell-free testing of the pSB3K3-pmntP-riboswitch-NanoLuc sensor response to MnCl2. (A) NanoLuc sensor dose response to MnCl2 with fold-change values calculated as the change in luminescence relative to the untreated (0mM MnCl2) control measured at 2, 4, 6 and 24hr. (B) Dose-response test NanoLuc sensor to MnCl2 at the 2-hour time point. Error bars in panel C indicate the mean ± 1 standard error.
Conclusions from Experiment 10:
- The assay yielded similar fold-changes in response to MnCl2 at timepoints from 2 hr to 24 hours, indicating that the test could be run as a rapid 2hr assay.
- The NanoLuc sensor responded to MnCl2 with a dose-dependent increase in luminescence from 0.01mM - 1mM MnCl2.
Experiment 11: Determine the limit of detection for the pSB3K3-pmntP-riboswitch-Nanoluc sensor response to MnCl2 in cell-free assay. The response of the Nanoluc sensor to an extended range of MnCl2 doses was measured to estimate the limit of detection for the assay (Table 8). A no DNA sample was included as a no plasmid DNA negative control (-C), and the inducible NanoLuc plasmid (pET-28a(+)::NL) served as a positive control (+C).
Table 8: Cell-free test of pSB3K3-pmntP-riboswitch-NanoLuc manganese sensor response to MnCl2. Reactions were set up in PCR tubes in 12.5µl volumes as indicated and incubated at 29C. At 2hr and 4hr, 3µl of this reaction was taken for measurement in 3 technical replicates (1µl sample + 9µl water + 10µl furimazine substrate) in a white plate using the NanoGlo Luciferase Assay kit.
Abbreviations: Nanoluc +C: pET28a(+)::NL, NanoLuc Sensor: pSB3K3-mntP-riboswitch-NanoLuc.
Results: The NanoLuc sensor generated luminescence in a dose-dependent manner in response to 0.0025 - 0.5mM MnCl2 (Figure 27A, adjusted R2 = 0.9949). The limit of detection (LOD) of the NanoLuc assay was calculated using the formula LOD = BLSD + 3SD control sample + 3 standard deviations (cite). We converted the LOD luminescence limit (677,380) to mM MnCl2 using the formula LOD(mM) = mx+ b = 1,135,245.46 * 677,380.25 + 595,434.71 = 0.007mM MnCl2. The LOD of 0.007mM MnCl2 corresponds to 0.4ppm. Luminescence of these samples was also measured in a 96-well white plate using an iPhone 11 (30 second exposure, ISO 8000, 26mm f1.8, 12MP) (Figure 27B), and the luminescence emitted was sufficient for imaging.
Figure 27: Dose response of the pSB3K3-pmntP-riboswitch-NanoLuc sensor to 0.0025 - 0.25mM MnCl2 by cell-free assay. (A) Dose-response was measured 2 hours after assay start and imaged on a BioTek Synergy H1 plate reader. Error bars indicate +/- 1 standard deviation. (B) Luminescence measured by iPhone imaging.
Conclusions from Experiment 11:
- The cell-free assay performed using the pSB3K3-pmntP-riboswitch-NanoLuc sensor demonstrated a limit of detection (LOD) of 0.007mM MnCl2 (0.4ppm). This LOD is below the 0.5ppm exposure limit suggested by O’Neal et. al. 2015 [3] and equal to the World Health Organization limit of 0.4ppm established in 2004 [4].
- The luminescence generated by the NanoLuc sensor is sufficient for imaging using a mobile phone.
Experiment 12: Demonstrate the capability of the NanoLuc sensor to detect manganese in water samples. Several raw and filtered water samples were obtained from grab samples taken as part of Dr. Stephen Jacquemin’s (WSU Lake Campus) routine Grand Lake St. Marys watershed monitoring work. These samples were frozen prior to receipt for testing in our assay. All samples were imaged using the Biotek Synergy H1 plate reader and the Luminescence Imaging Device (LID) hardware described in the Hardware Wiki page. The LID hardware is a 3D printed device that serves as a portable darkroom. It has a holder for a mobile phone and test samples to ensure consistent imaging from run to run.
Table 9: Cell-free test of water samples using the pSB3K3-pmntP-riboswitch-NanoLuc sensor. Reactions were set up in PCR tubes in 12.5µl volumes as indicated and incubated at 29C. At 2hr, 3µl of this reaction was taken for measurement in 3 technical replicates (1µl sample + 9µl water + 10µl furimazine substrate) in a white plate using the NanoGlo Luciferase Assay kit.
Abbreviations: -C: no plasmid DA (negative) control; Nanoluc +C: pET28a(+)::NL, NanoLuc Sensor: pSB3K3-mntP-riboswitch-NanoLuc.
Results: Measurements made using the BioTek Synergy H1 plate reader yielded a linear dose-dependent increase in luminescence in response to manganese from 0.01mM - 0.5mM (r2 = 0.99, Figure 28A). The linear regression of the standard curve was performed to calculate the manganese levels for the water samples (Table 10, “Plate Reader” data).
In parallel, manganese levels from the same reactions (Table 9) were measured using the Luminescence Imaging Device (LID) hardware and an iPhone 11 (described in the Hardware Wiki, Figure 28B). The resulting light image (bottom) shows all 8 sample tubes were clearly visible. The MnCl2 samples showed the expected dose dependent increase in luminescence (middle), clearly evident in the line profile trace (top) and linear regression (Figure 28C, r2 = 0.987).
Figure 28: Comparison of manganese measurements made on a plate reader and the LID device. A) MnCl2 dose curve measured on a BioTek Synergy H1 plate reader. (B) Images collected using the Luminescence Imaging Device (LID) hardware including a light image of the tubes in place, the luminescence image, and a pixel intensity trace of the luminescent image. (C) Standard curve determined from iPhone imaging measurement in panel B quantitated using ImageJ.
Manganese levels in the water samples #1003 (sample #1), #1004 (sample #2) from Mike Ekberg, Miami Conservancy District, and the cold water (sample #3) and wetland (sample #4) samples from Stephen Jacquemin, Wright State University Lake Campus (Table 9) were calculated using the standard curve in Figure 28C. The calculated mM levels were converted to ppm and are listed in (Table 10, LID).
Table 10: MnCl2 measurements from water samples made using a plate reader and the LID hardware. Plate reader: BioTek Synergy H1 plate reader; and LID: the luminescence imaging device and iPhone 11.
Conclusions from Experiment 12:
- As a proof of principle, we showed that all water samples tested exhibited manganese levels higher than the limit of detection for the NanoLuc assay.
- We observed that the ppm values measured using the plate reader and the LID varied were not exactly similar. However the Manganese levels measured in all samples were above the minimum allowable ppm concentration.
- Further optimization of the LID will be performed in the future.
- Based on these results the 3D printed LID hardware can serve as a fieldable alternative to more expensive plate readers.
Conclusion
Adapting the pmntP-riboswitch-based biosensor to a cell-free format and incorporation of a NanoLuciferase reporter has (1) reduced assay time to around 2hrs, (2) removed the use of live bacteria in the test for improved fieldability, and (3) improved the magnitude of the reporter response to manganese. Together, these improvements have provided a biosensor that can be used without the need for complex equipment, visualized and quantified with a simple 3D printed mobile dark room device that performs similarly to more expensive plate readers. These improvements have greatly enhanced the fieldability of the sensor.
Future Work
The team plans to continue this project into the 2024 iGEM season. We have plans to perform the following work:
- Test the specificity of the pmnt-riboswitch-NanoLuc sensor to manganese. Prior characterization of the pmntP-riboswitch suggests high specificity for manganese. We plan to confirm specificity by testing sensor function with chloride salts of other heavy metals (e.g. Mg, Cd, Fe, Co, Cu).
- Complete work on our mobile app for use with the Luminescence Imaging Device (LID) hardware and field-test our manganese detection approach. We have designed a portable, 3D printed device to make imaging of the cell-free testing results easier for the end user, and a mobile application (currently in prototype) to guide the user through the image acquisition process, perform QC checks to make sure that the assay was standard curve is linear and appropriate, and calculate the concentration of manganese in the water test sample(s). The 3D printed device and mobile phone application are described in detail on the Implementation page.
- We have developed a strategy to chelate manganese from drinking water using a manganese binding protein that has been shown to exhibit high specificity for manganese, which will be utilized in conjunction with our manganese biosensor. This strategy is described in the Implementation page.