Results | Stony-Brook

TEDA - Gel Electrophoresis

Below is a selection of gels relevant to our cloning, with explanations for what each meant. This includes all of our fragments, linearized plasmid backbones, and linearized plasmids that are partially assembled. Note that our PCR reactions have resulted in specific bands within the target length (except for our Ago2 fragments which we had to perform a gel extraction).

Dual Regulation

From left to right we have linearized PACYC plasmid backbone, GFP fragment (with promoter, ribosome binding site (RBS), and terminator sequence), the first linearized assembled PACYC plasmid with promoter+RBS+GFP+terminator, repressor fragment (with promoter, ribosome binding site, target miRNA 326 sequence after the start codon of LacI, rest of LacI, P2A, L7Ae, and terminator sequence), all with overhangs.

GST-Argonaute2

From left to right we have linearized PJL1 plasmid backbone (X2), Non-codon optimized Ago2 fragment (with promoter, ribosome binding site (RBS), and terminator sequence), first half of codon optimized Ago2 fragment (with promoter and ribosome binding site), second half of codon optimized Ago2 fragment (with terminator), MBP tag fragment, all with overhangs.

More Repressor fragments + Ago2 phase 2 fragments

From left to right we have repressor fragment (microRNA target site in the 5' UTR, overwriting a later portion of the ribosome binding site), first half of repressor fragment (microRNA target site in the open reading frame/coding sequence, buffered by an additional P2A), second half of repressor fragment (microRNA target site in the open reading frame/coding sequence), linearized assembled PJL1 plasmid with codon optimized Ago2, GST tag fragment, all with overhangs.

TEDA - Cloning and Sequencing Results

For each plasmid we cloned using TEDA or attempted to edit using site-directed mutagenesis, we would then purify that plasmid from the cloning vector and send it off to sequencing. We did this to gain confirmation on whether the cloning was successful or not. Below is a selection of the plasmids we sent off the sequencing and related commentary, including indication of whether we went on to use a given plasmid in future experiments.

Dual Regulation Plasmid

1. GFP Ordered in Small Fragments: Initial attempts, in which we attempted to assemble several small fragments, including a separate promoter, RBS+coding sequence, and terminator part, were unsuccessful. The plasmid did not uptake the small gene fragments.

2. GFP Whole Fragment: Primers were designed to insert a newly ordered whole GFP fragment, from promoter to terminator, into the pACYC backbone. Unlike the previous attempts with smaller fragments, this was successful in creating the GFP plasmid that we would go on to use for later experiments as the “control” for normal GFP expression.

3. Insertion of Dual-Regulation System: Primers were then designed to insert sequences for all three proposed versions of the dual-regulation system, this time also all designed from promoter to terminator to avoid the issues encountered with cloning small fragments in TEDA. The three plasmids show that the fragments were successfully cloned into the new GFP-pACYC backbones. This occurred with no issues for DR with the target site in the 5’ UTR and DR with the target site in the ORF followed by P2A. We are conducting characterization of those two plasmids and hope to present more information than is currently provided at the Jamboree. However, in our plasmid into which we cloned DR with the target site as a scar on LacI, there was a point mutation. Specifically, a deletion occurred of one guanine in a series of four guanines that constituted a region of overlap between the target site and the LacI coding sequence. As a result, the open reading frame for nearly the whole DR construct was displaced. We went on to test the former two DR plasmids, and attempted to correct the latter with site-directed mutagenesis.
See the images below for the three plasmids.

The three plasmids show that the fragments were successfully cloned into the new GFP-pACYC backbones. This occurred with no issues for DR with the target site in the 5’ UTR and DR with the target site in the ORF followed by P2A. We are conducting characterization of these two plasmids and hope to present more information than is currently provided at the Jamboree. However, in our plasmid into which we cloned DR with the target site as a scar on LacI, there was a point mutation. Specifically, a deletion occurred of one guanine in a series of four guanines that constituted a region of overlap between the target site and the LacI coding sequence. As a result, the open reading frame for nearly the whole DR construct was displaced. We went on to test the former two DR plasmids, and attempted to correct the latter with site-directed mutagenesis.

4. Site Directed Mutagenesis: We designed primers meant to re-add the missing guanine into the plasmid with DR with the target site as a scar. However, these mutagenies failed. Plasmidsaurus was unable to sequence any of the plasmids we purified from cells that were transformed with the PCR products of SDM. We considered it is possible due to higher-than-preferred annealing temperature that the SDM failed. Attempts to redo this SDM with different primers with more appropriate annealing temperatures are ongoing.

Ago2 Plasmid

1. Based on our experience with the smaller fragments in cloning GFP into pACYC, we ensured that our attempts to clone Ago2 into pJL1, which had MBP in its backbone, involved only large fragments. We designed appropriate primers to remove MBP and add Ago2. After this, we attempted our first transformation of Ago2 into pJL1 which was successful, aside from two non-frameshift mutations. The first was a silent mutation, and second was a missense mutation that changed a leucine to an isoleucine. As discussed on the part page for Codon Optimized GST-Ago2, we did not believe these mutations to be significant as the former did not change what amino acid was coded for and the latter changed one amino acid with a hydrophobic side chain to another. All the while, the latter was not located in a region considered significant to the core structure and function of the protein as far as we could ascertain. We proceeded to continue cloning with this construct. In addition, we attempted to construct a plasmid with an Ago2 gene that was not codon optimized as a potential alternative option. This was not successful as none of the sequenced plasmids had successfully taken up Ago2, and some had not even removed MBP. Due to this only being a potential alternative, we abandoned further attempts to clone non-codon optimized Ago2.

2. We then needed to add a tag 5' to the Ago2 coding sequence. A GST tag to add was generously provided to us by the Glynn lab at Stony Brook. We also attempted to re-add MBP in the 3' region. We designed primers and attempted cloning which was partly successful. In the construct meant for MBP to have been added, cloning was unsuccessful. Due to time constraints, we decided to focus on the construct to which we added GST, which was successful. In this construct, no further mutations occurred. We proceeded to conduct characterization with this construct.

Confocal Microscopy

As we hope to detect microRNA concentrations in patients' bloodstream, we have to consider pertinent regulations regarding handling of blood. The United States Food and Drug Administration (FDA) maintains several pertinent regulations with regards to blood collection, preparation, storage, transportation, and testing. Upon review of these regulations, we found that outside of the blood handling, storage, transportation, and disposal performed by clinicians, our system will need to conform with FDA regulations regarding In Vitro Diagnostic products. This includes both for our actual cell-free system once it receives microRNA, as well as the microfluidic assay from which the miRNA will be extracted from the blood.

Initial miRNA Uptake Trial

The initial trial for miRNA uptake, observed through the confocal microscope at 60x optical focus, yielded promising but inconclusive results. The trial exhibited significant internal noise, compounded by high cell density, as evidenced by bright fluorescent patches in all samples, including the negative control. Despite this, the results qualitatively suggested some level of miRNA interaction with the cells, particularly in the 1 nM concentration group. However, the fluorescence did not scale linearly with increasing miRNA concentrations, indicating additional experimental errors.

Troubleshooting

We hypothesized that the internal noise might have been caused by the tagged miRNA adhering to the external membrane of the cells rather than entering them. The inconsistency in the scaling between fluorescence and miRNA concentration could have resulted from variations in cell concentration between samples. To address the first issue, we decided to thoroughly clean the samples by adding two additional PBS washes to eliminate any extraneous fluorescent signals. To ensure consistent cell counts across samples, we verified optical density using a plate reader and conducted serial dilutions of the known optical density samples to standardize the cell concentration per sample.

Successful miRNA Uptake Trial

After implementing the changes identified during troubleshooting, we conducted a second miRNA uptake trial using 100x optical magnification. This trial included a slide with 1 nM of miRNA and a control with no miRNA. The results showed significantly less internal noise, suggesting that the additional washing steps improved the accuracy of the miRNA uptake imaging. The images clearly demonstrate that miRNA at a 1 nM concentration successfully enters our cell strain. Further experiments with higher confocal magnification and picomolar miRNA concentrations could provide additional support for our hypothesis that blood-level concentrations of miRNA can effectively enter and interact with our cell system.

GFP Production

Although we successfully generated quantitative data for GFP fluorescence, we opted to confirm these results qualitatively, as this aligns with how the test is likely to be performed in practice. The confocal microscopy results were quite clear, showing minimal internal noise and a distinct difference in fluorescence between the GFP-producing plasmid and the bacteria without the plasmid. These findings are promising for the future implementation of Micronaut as a qualitative test. We will try to amplify fluorescence signals through testing other variants such as sfGFP. By doing so, we hope to reduce the difficulty of fluorescence measurement in our system so that it requires minimal imaging equipment.

Cell Plate Reader Results

To test the efficacy of our system, we quantified the fluorescence emitted under various conditions: Nothing, empty PACYC backbone, PACYC plasmid with just promoter+RBS+GFP+terminator, and PACYC plasmid with GFP and the dual repressors (LacI and L7Ae). Due to limitations to our equipment, we can only measure fluorescence emitted with an excitation wavelength of 485 and emission wavelength of 535 (as opposed to the ideal wavelengths of 475/509) (To test the efficacy of our system, we quantified the fluorescence emitted under various conditions: Nothing, empty PACYC backbone, PACYC plasmid with just promoter+RBS+GFP+terminator, and PACYC plasmid with GFP and the dual repressors (LacI and L7Ae). Due to limitations to our equipment, we can only measure fluorescence emitted with an excitation wavelength of 485 and emission wavelength of 535 (as opposed to the ideal wavelengths of 475/509) (Bauer et al., 1996).

1. Categorization in Bacteria

Our first initial measurement of fluorescence yielded unexpected results as our negative control group (bacteria with no plasmid transformed or just an empty PACYC backbone with no GFP protein) had higher fluorescence than bacteria transformed with our plasmid containing just GFP (groups A,B, and C). We think that this is due to the cell density of the bacteria being too high, and signals being blocked off and not properly read by the plate reader. Thus, we performed a serial dilution of the above colonies to find the optimum cell density to measure fluorescence within our system.

After dilution to an OD of 0.5 and below, our data relationship falls within what we expected; however, we were concerned as to whether GFP was expressed at all in groups transformed with the GFP plasmid due to the low fluorescence reading. To confirm our GFP expression, we checked our cells under the confocal microscope, as mentioned above. The microscopy result shows that there is a difference in fluorescence between MRE 600 cells with no plasmid and MRE 600 cells with our GFP plasmid. This preliminary data suggests that GFP expression was too low to be detected in our bacterial system with a plate reader. We believe that the expression level of GFP is not high enough due to 1) the protein being constitutive expressed and the condition (incubation time and temperature was not optimal to produce enough GFP for detection by plate reader 2) there were too many barriers and noise such as the cell membrane and cell wall which further hinders the detection of fluorescence signal using a plate reader.

Due to limitation in our available lab space, time, and equipments, we decided to proceed with testing our system within a cell-free system in attempts to solve issue #2.

2. Categorization in Cell-Free System

Since our system is under the endogenous E. coli promoter as opposed to T7, we tested the different incubation times to optimize protein expression within the range recommended by NEB for their S30 cell-free system for genes under T7. Our initial data shows an increase in fluorescence with cell-free systems incubated with a plasmid containing GFP vs.just the empty backbone or deionized water at 4 hours and 6 hours.

We tested our system with a dual repression system (where the microRNA target site is placed in the open reading frame/coding sequence, buffered by an additional P2A). There is significant decrease with the dual repressors; however, there is still a high background fluorescence with just the PACYC backbone. We will continue to investigate this further before the jamboree, but we believe that the high background does not take away that significance change after the addition of the dual repressors.

Finally, we categorized our full circuit consisting of DR (Placing the microRNA target site in the open reading frame/coding sequence, buffered by an additional P2A), Ago2, and our target miRNA (miR-326). This preliminary data illustrates that without miRNA added, the fluorescence stays at the level suppressed by DR. The addition of miR-326 increases the fluorescence; however, further testing needs to be done to ensure this is reproducible, to understand why the addition of miR-23 also increased the fluorescence, and whether this will continue to occur.

Conclusion - Proof of Concept

When we developed our project proposal, we hoped to create a system that could detect miRNA using fluorescence as a method of diagnosing B-cell lymphoma. We set out three goals for our system, and below is a synthesis of these goals and our progress towards them.

1. Without any repression, GFP expresses:

Our preliminary results indicate that there is a fluorescence signal in both bacteria and cell-free systems expressing GFP under the endogenous e.coli promoter. Further testing and troubleshooting needs to be done in regards to the high background we observe.

2. With DR, GFP no longer expresses.

Fluorescence signal decreases significantly with the addition of dual repressors as indicated by the plate reader results. We were only able to test one of three approaches we designed. Preliminary cell-plate reader evidence shown above also shows recovery of expression after Ago2, suggesting a complete circuit.

3. With Ago2 and miRNA-326, GFP expresses once again

Our preliminary attempts at expressing Ago2 within our bacteria system depicts that further optimizations are needed. As we know the gene was successfully cloned into the expression vector, we would like to try different conditions such as a lower temperature and longer incubation time to acheive expression. We would also like to perform a Western Blot for Ago2 produced in the S30 cell-free system to observe whether or not Ago2 expression is sucecssful in that context.

Overall

Our preliminary results illustrate our system's potential to be a functional screening test, but equipment such as a plate reader is needed in its current form. In our discussion with various stakeholders, a common recurring theme is the importance of our system being easy to use and having low equipment requirements. We wish to remove the need of this expensive equipment through the following changes that will amplify our fluorescence signal:
1. Switching from endogenous e.coli promoter to T7 Promoter.
2. Using other GFP variants such as sfGFP, or other genes whose expression is visible to the naked eye.
3. Implementation in a paper-based cell-free system, as proposed by 2022 ICJFLS iGEM Team for their miRNA-detecting system ("Proposed Implementation", 2019).