Measurement

Introduction

---

We have designed our experiments with reproducibility, accuracy and reliability in mind; to achieve this we ensured use of appropriate controls to verify that our Cas13a system works as intended and to test the components and models used. We used excitation and emission spectral scanning to determine the optimal excitation and emission wavelengths that would provide the best results for us. We also have converted all of our results from arbitrary fluorescence to fluorescein concentration equivalents (absolute values) to allow future iGEM teams to easily interpret our data and be able to ensure consistency of experimental values which is not possible using relative measurements such as arbitrary fluorescence values. Detailed protocols are also provided as well as tools to assist in the designing of specific experiment to allow for other iGEM teams to reproduce our system effectively. We have aimed to provide results in an unambiguous way such as providing all R2 values from the models we have used in the data as well as providing data that could be graphed as a bar chart as scatter plot with bars as to not hide any data points.

Optimising Wavelengths Used in Fluorescent Spectrophotometry

---

First, we used varying concentrations of fluorescein and DNaseAlertTM probes to run an excitation and emission scan using a Tecan (Infinite 200 Pro) to determine the wavelengths we would use in our further experiments. This was necessary as our spectrophotometer does not have bandpass filters, this meant the recommended excitation and emission wavelengths would produce spillover and therefore not give accurate results.

[Figure 1] Figure shows the excitation and emission spectra of fluorescein (RNaseAlert probe fluorophore) and the DNaseAlert probes (fluorophore is hexachlorofluorescein). No replicates were done due to no biology being in the system and therefore it would only be a technical replicate providing no value to the result.

The Tecan (infinite 200Pro) required at least 30 nm between excitation and emission wavelengths being used and this therefore restricted wavelengths available to use. For the wavelengths used with fluorescein experiments (including RNaseAlertTM probe experiments, due to having fluorescein as the fluorophore in the probe) was 488 nm excitation and 519 nm emission. Then for DNaseAlertTM probe experiments we concluded that 535 nm for excitation and 565 nm emission was the best wavelengths to choose. These wavelengths were chosen by using data from Figure 1 to avoid using wavelengths that would cause an increase in fluorescence even when no fluorescent sample is in the well; this is called spillover. Avoiding spillover was a really important issue for us as spillover could introduce false positives. From some of the interviews done with famers such as David Andrews from Warsons Beef, our human practices told us that false positives are particularly detrimental due to having to unnecessarily cull their herd which can have highly adverse impacts on their livelihood. Therefore, minimising the probability of false positives was one of our main priorities.

Fluorescein Concentration Curves

---

Fluorescein was chosen as a standard because the fluorophore in the RNaseAlert^TM probe is fluorescein ^[1], allowing our standard to have the same extinction coefficient (13700 M-1cm-1 [2]) as our probe, but also the same excitation and emission spectra as our RNaseAlertTM probes implemented in the Cas13a system allowing for the standard to be very accurate. Also, as fluorescein is included in the iGEM ‘Fluorescence Measurement Calibrants Kit’ this allows all other iGEM teams, even with limited budget, to follow our protocol increasing accessibility. The "Fluorescein Concentration Curves protocol (see below) used allows other iGEM teams to make fluorescein concentration curves, allowing for conversion of relative values to their own absolute values. For all other protocols please see Experiments Page

We maintained the same variables as the fluorescein concentration curve for each experiment including temperature, volume of solution and excitation and emission wavelengths; this is important as a change in any of these variables would results in a change in the relative values acquired through fluorescence measurements.

[Figure 2] A graph that shows the variation in fluorescence measurements against fluorescein concentration for different gains on the Tecan. Gain 10 has a low R squared value because low gains have lower resolution.

Figure 2 shows the graphs other iGEM teams should produce when repeating this experiment at multiple gains. Changing the gain changes the amplification of the signal and can strongly influence fluorescence measurements. If the output signal is low, gain can be increased to amplify the signal to allow for detection by the detector. However, if the output signal is high gain can be reduced to prevent saturation of the detector. By producing graphs of fluorescence versus increasing fluorophore concentration at multiple gains, the results can be used to standardise fluorescence data in a variety of experimental conditions.

[Figure 3] Michaelis-Menten graph for varying concentration of RNaseAlert probe

We then wanted to test which of our gains would provide the most reliable and accurate results, to do this we needed to test various concentrations of our probe with a fast acting ‘model’ enzyme, RNaseA. Figure 7 shows increasing fluorescence (in fluorescein concentration equivalents, see standard curve below) with time at several RNA probe concentrations. The curves were fitted to the Michaelis-Menten equation and the R2 values shown. Even when the RNA probe concentration was at the maximum value allowed for by the kit (500 nM), the highest fluorescein concentration equivalent was 60 nM. This falls into the range of the standard curve measured at gain 100. A gain of 100 therefore gives the best range for our experiments and as it is a higher gain, it will have a higher accuracy for reporting our results in.

[Figure 4] A graph showing fluorescence against fluorescein concentration at a gain of 100 on the Tecan.

Figure 4 shows the graph of fluorescence versus increasing fluorescein concentration at our chosen gain of 100. The graph is fitted to a linear equation which was then used to transform all of our data into fluorescein concentration equivalents.

One limitation of using the fluorescein concentration curves is for any iGEM team using filter wheel-based fluorescence reader may not be able to get to the same excitation and emission wavelengths. Therefore, other iGEM teams using filter wheel based fluorescence readers will not be able to directly compare to our results however, these teams will be able to compare results between each other while using the same wavelengths.

Our use of Unambiguous Data and Other Absolute Values

---

[Figure 5] A scatter plot with bar chart, showing the % change in fluorescein concentration equivalents compared to Blood + fluorescein (10 nM). The controls are: PBS + fluorescein (10 nM), PBS, and Pigs Blood.

In Figure 5 and other graphs for reducing blood opacity (results link) we decided to plot the graphs in a scatter plot with bar chart to avoid hiding any of the data points and provide unambiguous results. We converted all data to fluorescein concentration equivalents for an absolute value for easier interpretation, but the data still provided unclear results as it was difficult to see how much the treatments used would increase the fluorescence in the samples. Therefore, we decided to use percentage changes compared to blood + fluorescein (10 nM) as a baseline to allow stakeholders, end users and other scientists to more easily see which treatments provided the largest increase of fluorescence. The negative and positive aspect of the graph allows very easy assessment of whether the treatment has increased or decreased the fluorescence intensity of the sample.

Quantifying Cas13a Purity from Analysis of SDS-Page Image

---

To quantify purity of Cas13a, the following analysis was used:

• The image was cropped to the relevant lanes, (shown in yellow on Figure 7) with the band containing the Cas13a protein also separately cropped (shown in red in Figure 7).
• These images were saved as .png files named img1-4.
• A custom Matlab script shown below was used to import each image, convert to a numerical matrix of RGB intensities, and count blue pixels given a certain threshold (Red <=100, Green <=100, Blue >= 100).
• Final values were plotted [Figure 6]

[Figure 6 and 7] 6: Bar graph of %Purity for His Trap and SEC. No scatter dots are on the graph as there is only one value for the analysis of the data. 7: An image of resulting SDS-page gel after AKTA Pure SEC and His trap column, surrounded by yellow box. The red box highlights specific bands.

See below the Matlab script for quantifying Cas13a purity from analysis of SDS page image.

How Controls are used Throughout our Experiments

---

Throughout our experiments controls have always been very important to validate the results we were seeing, and this was even more prevalent for our test of our Cas13a system involving the transcribed sgRNA and targets, the expressed and purified Cas13a and validated probes. We needed to ensure that our system required all of the components to verify our understanding of the mechanics behind our system. To do this we set up systematic controls to remove each component to guarantee there was no activity. We also wanted to test the accuracy of our model for the design of the sgRNA, to do this we incorporated a target where the sgRNA designed was shown on the model to have multiple interactions which were presumed to hinder the binding affinity to the cas13a and therefore inhibit the reaction and fluorescence produced. Furthermore, we wanted to test the use of the wrong target to check our system would only be targeting our specific targets, this was to ensure no false-positives as through our human practices with farmers they have stated that false-positives lead to culling of livestock and therefore big loss of livelihood. All the controls used in the testing of our Cas13a system are shown in Figure 8.

[Figure 8] Shows six graphs of fluorescein concentration equivalents against time each graph has a different control. This enables us to confirm and validate each individual component of our diagnostic system. PLAUR sgRNA/target plots follow a Michaelis-Menten model with the real data plotted as error bars. There was only 2 biological replicates for each data set due to budget limitations.

[Figure 9]Michaelis-Menten graph for varying concentration of RNaseAlert probe.

Figure 9 shows the validation of our RNaseAlert probe and acts as a positive control for our RNaseAlert probes in the cas13a system test.

All of our data will be in absolute units and all experiments will be done with at least 3 biological replicates and any exceptions due to time or budget will be explicitly stated in figure legends. Error bars will always be reported in standard deviations.

Future Work

---

We wanted to find a better standard for the DNaseAlertTM probe as the fluorophore used is hexachlorofluorescein (HEX) [3] which has different excitation and emission spectra as well as different extinction coefficient (31600 M-1cm-1 [4]) to fluorescein. HEX was planned to be our other standard however, HEX proved difficult to buy and so we are currently working on a cheap synthetic route to HEX as shown in Figure 10 that in the future could be adopted to allow future iGEM teams to achieve a better standard.

[Figure 10] HEX pathway made on Chemdraw designed from^[5] and ^[6]