An Accurate, Fast, and Simple Viral Diagnostic Test
TAS_Taipei 2020
Team
iGEM Student Team Members:
Wilson Huang, Hannah Hsu, Joyce Ting, Jimmy Su, Matthew Fang, Tsuyoshi Misaki, Derek Chan, Justin Yang, Ting-Yu Yeh, Kelly Yang, Vera Chien, Tiffany Huang, Andrew Chen, Claire Wei, Tyler Chen, Renee Chien
iGEM Team Primary PI:
Jude C. Clapper
iGEM Team Secondary PI:
Jonathan Hsu
iGEM Team Advisor:
Nicholas J. Ward
Abstract
Seasonal flus and pandemics, which account for millions of infections and hundreds of thousands of deaths, require rapid and reliable detection mechanisms to implement preventive and therapeutic measures. Current detection methods of viral infections have limitations in speed, accuracy, accessibility, and usability. This project presents a novel, widely applicable viral diagnostic test that uses a modified version of rolling circle amplification (RCA) to be sensitive, specific, direct RNA targeted, colorimetric and operable at room temperature. We are specifically detecting the following high-impact viruses: SARS-CoV-2, Influenza A (H1N1pdm09), and Influenza B (Victoria Lineage), although our test can be adapted to any viral infection. Results using synthetic viral DNA and RNA sequences show that our diagnostic test takes approximately one hour, detects femtomolar concentrations of RNA strands, and differentiates between virus strains. We believe implementing our diagnostic test will provide faster responses to future viral-related outbreaks for quicker societal recovery.
Team:TAS Taipei/Poster
Poster: TAS_Taipei
Figure 1: Influenza Cases Worldwide March 2020 (WHO | Influenza Update - 362, n.d.).
Seasonal flus (Figure 1) account for approximately 3 to 5 million cases and 250 to 500 thousands deaths worldwide annually (Influenza (Seasonal), n.d.). Pandemics such as the ongoing COVID-19 pandemic (Figure 2) has already caused 30.9 million cases and 958 thousand deaths, and the numbers continue to rise (WHO Coronavirus Disease (COVID-19) Dashboard, n.d.). The great quantity of people subjected to these diseases makes diagnostic measures crucial to installing preventive and therapeutic action.
Figure 2: Worldwide SARS-CoV-2 Cases Up to Oct 23rd, 2020 (Microsoft Bing COVID-19 Tracker, n.d.).
References
Influenza (Seasonal). (n.d.). Retrieved September 21, 2020, from https://www.who.int/news-room/fact-sheets/detail/influenza-(seasonal)
Microsoft Bing COVID-19 Tracker. (n.d.). Retrieved October 25, 2020, from https://www.bing.com/covid?ref=share
WHO | Influenza update—362. (n.d.). WHO; World Health Organization. Retrieved October 25, 2020, from http://www.who.int/influenza/surveillance_monitoring/updates/2020_03_02_update_GIP_surveillance/en/
WHO Coronavirus Disease (COVID-19) Dashboard. (n.d.). Retrieved September 21, 2020, from https://covid19.who.int
Figure 1: Human Practices Interviews on Test Kit Experts
Through various expert interviews with Ms. Pearl Fong from United BioPharma, Dr. Michael Jewett from Northwestern University, and Dr. Chang Sui-Yuan from NTU, and even conferences on Taiwan Biotech Solutions (Figure 1), we obtained useful information to research the uses of various test kits in the market. Upon further research, we were able to identify the advantages of disadvantages of 3 main methods: molecular, antibody, and antigen tests (Giri et al., 2020)(Figure 2).
Figure 2: Different Targets in the Modes of Viral Diagnosis
Antigen and antibody tests through ELISA may have high throughput, but the need of an immune response often leads to delayed diagnoses (Giri et al., 2020) . Although lateral flow immunoassays are quick, their low sensitivity results in a high rate of false negatives (Gordon & Michel, 2008). Molecular testing is the most sensitive and specific test type but requires highly technical instrumentation and operation (Corman et al., 2020).
Figure 3: Nasopharyngeal Swab from Miraclean Technology (Nasopharyngeal Sampling Swabs, n.d.)
As for viral sample collection, nasopharyngeal swabs, which require medical personnel, safety equipment, and risk healthcare workers, tends to be used(Iwasaki et al., 2020)(Figure 3). As for nucleic acid isolation for PCR tests, RNA Extraction kits that require membrane columns, buffers, and centrifugation are used.
Figure 4: Comparison of Molecular (PCR), Antibody, and Antigen Test accuracy across time of infection
Therefore in our project, we decided to employ a molecular testing method due to its high accuracy while mitigating its downsides through various modifications. Our hardware allows patients to collect test samples themselves using their saliva. Thus, our viral spiral test kit was created to give sensitive and specific results that can be determined at home.
References
Corman, V. M., Landt, O., Kaiser, M., Molenkamp, R., Meijer, A., Chu, D. K., Bleicker, T., Brünink, S., Schneider, J., Schmidt, M. L., Mulders, D. G., Haagmans, B. L., van der Veer, B., van den Brink, S., Wijsman, L., Goderski, G., Romette, J.-L., Ellis, J., Zambon, M., … Drosten, C. (2020). Detection of 2019 novel coronavirus (2019-nCoV) by real-time RT-PCR. Eurosurveillance, 25(3). https://doi.org/10.2807/1560-7917.ES.2020.25.3.2000045
Giri, B., Pandey, S., Shrestha, R., Pokharel, K., Ligler, F. S., & Neupane, B. B. (2020). Review of analytical performance of COVID-19 detection methods. Analytical and Bioanalytical Chemistry. https://doi.org/10.1007/s00216-020-02889-x
Gordon, J., & Michel, G. (2008). Analytical Sensitivity Limits for Lateral Flow Immunoassays. Clinical Chemistry, 54(7), 1250–1251. https://doi.org/10.1373/clinchem.2007.102491
Iwasaki, S., Fujisawa, S., Nakakubo, S., Kamada, K., Yamashita, Y., Fukumoto, T., Sato, K., Oguri, S., Taki, K., Senjo, H., Hayasaka, K., Sugita, J., Konno, S., Nishida, M., & Teshima, T. (2020). Comparison of SARS-CoV-2 detection in nasopharyngeal swab and saliva. MedRxiv, 2020.05.13.20100206. https://doi.org/10.1101/2020.05.13.20100206
Nasopharyngeal Sampling Swabs. (n.d.). Miraclean Technology Co.,Ltd. Retrieved October 25, 2020, from /nasopharyngeal-sampling-swabs-15936872289228819.html
RCA Theory
We created single stranded DNA segments (padlock probes) that act as our virus detector while our synthetic viral DNA and RNA mimics actual virus sequences (Figure 1). We specifically selected SplintR Ligase, which ligates DNA and RNA complementary hybrids, and Φ29 DNA polymerase, which displaces and synthesizes continual DNA strands, to run our reaction. We incorporated a pH indicator which allows us to visualize a color change through a change in hydrogen ion concentration. We also removed all buffers that might neutralize the hydrogen ions produced. The SplintR ligase enzyme that ligates the padlock probe enables high specificity through direct RNA targeting. SplintR ligase along with Φ29 DNA polymerase allow our reaction to operate at room temperature.
Figure 1: The rolling circle amplification RCA process.
Padlock Probe Design
Figure 2: Alignment of SARS-CoV-2, SARS, MERS across various cases and dates
We chose a 36 nucleotide target sequence that is conserved between various mutations within SARS-CoV-2 strains, but also exhibiting differences from SARS and MERS. We then used this sequence as our synthetic viral target and utilized it to design our padlock probes.
References
Hamidi, S. V., & Ghourchian, H. (2015). Colorimetric monitoring of rolling circle amplification for detection of H5N1 influenza virus using metal indicators. Biosensors and Bioelectronics, 72, 121–126. https://doi.org/10.1016/j.bios.2015.04.078
Hamidi, S. V., Ghourchian, H., & Tavoosidana, G. (2015). Real-time detection of H 5 N 1 influenza virus through hyperbranched rolling circle amplification. The Analyst, 140(5), 1502–1509. https://doi.org/10.1039/C4AN01954G
Hamidi, S. V., & Perreault, J. (2019). Simple rolling circle amplification colorimetric assay based on pH for target DNA detection. Talanta, 201, 419–425. https://doi.org/10.1016/j.talanta.2019.04.003
Our parts collection BBa_K3352000 to BBa_K3352009 gives us and anyone the ability to run RCA reactions at room temperature to detect any nucleic acid sequence of interest.
SplintR Ligase
SplintR ligase ligates gaps and nicks on the DNA sequence when a DNA and RNA hybrid is complementary (Figure 1).
Figure 1: Function of SplintR ligase
Φ29 DNA polymerase
From the 5’ to 3’ direction, Φ29 DNA polymerase displaces the existing strand with a helicase-like activity and synthesizes new DNA strands (Figure 2). From the 3’ to 5’ direction, it hydrolyzes and cuts off nucleotides.
Figure 2: Function of Φ29 DNA polymerase
Designs
We designed ORFs of SplintR ligase and Φ29 DNA polymerase sequences. We attached them to 6x histidine tags through glycine-serine linkers for purification purposes. Each composite part that we created contained different parts that helped us improve protein expression.
BBa_K3352008 & BBa_K3352009
We used pET vector parts that include an extended RBS with a long untranslated region UTR upstream and a T7 terminator (Figure 3). We were finally able to fully maximize the expression and isolate our two enzymes.
Figure 3: SplintR ligase (top) and Φ29 DNA polymerase (bottom) constructs using pET3a and pET11a plasmid parts respectively.
Protein Expression and Purification
We expressed our enzymes through BL21(DE3) E.coli cell liquid cultures. These were then diluted, lysed and purified through nickel affinity chromatography (Figure 4). To check if our proteins were expressed, we ran our eluate through SDS-PAGE protein gels to confirm the bands.
Figure 4: SDS-PAGE results on purified pET vector SplintR ligase(left) and Φ29 DNA polymerase(right). The expected size for SplintR ligase is 35.7 kDa and the expected size for Φ29 DNA polymerase is 68.2 kDa.
Figure 1: Masks are created by drawing polygons around the pixels representing the reaction. The masks are then applied to each frame, and the values of the isolated pixels are obtained and plotted against time.
We developed our own pH measurement software, VisualpH, that measures pH from a photo or video of a solution with a color based pH indicator (Figure 1). It’s not only fast and accurate, but it only uses relatively inexpensive and common tools. It quickly converts a video of a pH solution into a tensor of measured pH, which can be used in further analysis. Our software helped us measure the pH precisely and use that data for further analysis, whether that is for modeling, specificity or determining improvements to be made to methodology (Figure 2).
Figure 2: Sensitivity Test on our SARS-CoV-2 viral probe (C-19) quantified using VisualpH
Its usefulness however, is not limited to our own project. It could be used to quantify data from anything colour related, such as the activation of genes like mCherry or GFP, or maybe the properties of certain organisms or environments. This tool has an extensive range of possible applications, and it has many advantages over the current meta of options.
VisualpH works by first receiving a user defined mask that defines where the solutions are inside the image or video. It then compiles the hue data, and converts it into pH data.
Proof of RCA Theory
In order to confirm the validity of our RCA theory, we ran several RCA tests with DNA padlock probes and synthetic viral DNA targets. To a reaction tube we added the synthetic viral DNA target, its complimentary DNA padlock probe, Φ29 DNA polymerase, SplintR ligase and a pH indicator in a reaction solution. In theory, the target and the padlock probe would fuse, creating H+ ions that would rapidly lower the solution’s pH value. Our results showed a color change from pink to yellow in the tube containing the synthetic viral DNA target, suggesting that our RCA test was successful (Figure 1).
Figure 1: The results of the initial RCA test. The tube containing the viral target changed color which matched our predictions.
Confirmation of DNA Amplification
We wanted to confirm that DNA amplification was what was responsible for the color change instead of other factors. Therefore, we ran RCA reactions where one sample contained a positive target and another contained no target. Our results show a larger amount of DNA amplified when we added a positive target (Figure 2A). In addition, when we ran a DNA gel of the tests 30 and 40 minutes into the RCA reaction, the well containing the 40 minute sample was visibly brighter, thus suggesting that the DNA amplification has occurred rapidly (Figure 2B).
Figure 2: A: We ran 2 RCA reactions that were later transferred to DNA gels. The well on the left contained the reaction with no target (-) and the well on the right contained the reaction with a target (+). B: RCA reactions were stopped 30 and 40 minutes into the reaction to confirm DNA amplification (+). The samples were compared to reactions with no targets (-).
Since synthetic viral DNA targets worked perfectly for the RCA test, we switched to using synthetic viral RNA targets for all future tests to simulate actual RNA viral strands.
Sensitivity
We also wanted to determine how low of a target concentration our test can accurately detect, which is also known as sensitivity. We diluted the SARS CoV-2 synthetic viral RNA target from micromolar to attomolar to test our RCA under different concentrations. Our results demonstrated that a color change occurs in all concentrations until the attomolar concentration was reached (Figure 3).
Figure 3: SARS-CoV-2 RCA test using SARS-CoV-2 padlock probes. Different concentrations of synthetic SARS-CoV-2 targets were utilized.
Specificity
Finally, we determined how accurately our RCA test could distinguish between different virus strains, which is also known as specificity. Each padlock probe only matched with one specific target sequence, and our results show that a mismatched target had a differentiable rate of color change than a matched target (Figure 4). Thus, we determined that our test was specific to the virus type.
Figure 4: Influenza A RCA test using Influenza A padlock probes. The blue reaction has no RNA target, the orange reaction has an Influenza B target (mismatched to probe), and the green reaction has an Influenza A target (matched to probe).
References
Hamidi, S. V., & Ghourchian, H. (2015). Colorimetric monitoring of rolling circle amplification for detection of H5N1 influenza virus using metal indicator. Biosensors and Bioelectronics, 72, 121–126. https://doi.org/10.1016/j.bios.2015.04.078
Hamidi, S. V., Ghourchian, H., & Tavoosidana, G. (2015). Real-time detection of H 5 N 1 influenza virus through hyperbranched rolling circle amplification. The Analyst, 140(5), 1502–1509. https://doi.org/10.1039/C4AN01954G
Hamidi, S. V., & Perreault, J. (2019). Simple rolling circle amplification colorimetric assay based on pH for target DNA detection. Talanta, 201, 419–425. https://doi.org/10.1016/j.talanta.2019.04.003
The purpose of our mathematical model is to approximate the viral load of SARS-CoV-2 within a patient based on the reaction rate of our rolling circle amplification (RCA) detection test. Although viral load does not necessarily correlate with disease severity, it is an appropriate measure for the risk of transmissibility (Lescure et al., 2020). Through the process of constructing our mathematical model, we realized the effects of initial pH on our detection test and the activity of Φ29 DNA polymerase (Figure 1). However, from our later trials with a controlled pH, we noticed that there was a consistent difference between the rate of reactions containing different amounts of viral RNA (Figures 2, 3)We also discovered through an analysis of our regression function that the concentration of viral RNA can be modeled by an exponential function in terms of the pre-exponential factor from our regression function (Figure 4). Based on the assumption that one virus particle contains at least one strand of viral RNA, our model, when implemented into our software, allows users to easily approximate the minimum viral load within their system via a time-lapse recording of their RCA reaction(Figure 5).
Figure 1: The scatterplots for the reactions containing 0.0625 μM of target RNA in the first and second trials. The graph displays the concentration of hydrogen ions in micromolar over time. The separation of the plots for each trial suggest that the overall reaction rate is dependent on the initial pH of the reaction.
Figure 2: The regression lines for reactions containing 0.0625 μM and 0.000625 μM of target RNA in the three new trials. The general regression function is displayed in the top left corner.
Figure 3: The combined scatterplots for reactions containing 0.0625 μM (red), 0.00625 μM (orange), 0.000625 μM (yellow), and 0.0625 nM (purple) of target RNA in the three new trials. The graph displays the concentration of hydrogen ions in micromolar over time. Though the cluster is weaker for data of reactions containing 0.00625 μM and 0.000625 μM of target RNA, there is still an expected separation in data points for reactions of different target RNA concentrations.
Figure 4: Exponential regression for ‘a’ to target RNA concentration data. Because all parameters of the regression function are related to the pre-exponential factor ‘a’, the target RNA concentration can be approximated by this exponential function relating ‘a’ to target RNA.
Figure 5: Flowchart of sample collection to viral load analysis.
References
Lescure, F.-X., Bouadma, L., Nguyen, D., Parisey, M., Wicky, P.-H., Behillil, S., Gaymard, A., Bouscambert-Duchamp, M., Donati, F., Le Hingrat, Q., Enouf, V., Houhou-Fidouh, N., Valette, M., Mailles, A., Lucet, J.-C., Mentre, F., Duval, X., Descamps, D., Malvy, D., … Yazdanpanah, Y. (2020). Clinical and virological data of the first cases of COVID-19 in Europe: A case series. The Lancet Infectious Diseases, 20(6), 697–706. https://doi.org/10.1016/S1473-3099(20)30200-0
Figure 1: An isometric view of the sputum sample collection and purification device.
Our hardware is a sputum-based virus sample collection device (Figure 1). Combined with our universal virus detection kit, Viral Spiral, our device increases the practicality of at-home virus testing as it makes sputum sample collection and purification more accessible. The device comes prefilled with a solution at pH 10 composed of a lysis buffer, amine-functionalized diatomaceous earth (ADEs), homobifunctional imidoesters (HIs), and protector RNase inhibitors. Users transfer sputum into the device, cap the barrel with the plunger and thoroughly mix the sample. During this step, pathogens are lysed, and nucleic acids are attached to the surface of ADEs through electrostatic interactions assisted by HIs (Zhao et al., 2019). After 15 minutes, users pass the solution through a filter, trapping the ADE-nucleic acid complexes. Users then detach the filter and draw up an elution buffer at pH 10 which separates the nucleic acid from ADEs and HIs (Zhao et al., 2019). Lastly, the filter is reattached and the eluent is collected for analysis by our kit.
References
Zhao, F., Lee, E. Y., Noh, G. S., Shin, J., Liu, H., Qiao, Z., & Shin, Y. (2019). A robust, hand-powered, instrument-free sample preparation system for point-of-care pathogen detection. Scientific Reports, 9(1), 16374. https://doi.org/10.1038/s41598-019-52922-y
Through our research, we realized there are a lot of social implications regarding COVID-19 and test kits. To further the discussions, our team hosted a roundtable panel to discuss the ethics of COVID-19 in March (Figure 1). On one hand, we wanted to provide a space where people can share their opinions, and on the other hand, we used the feedback to further shape our approach to integrated human practice.
Figure 1: The team led the discussions for the bioethics panel
From our Bioethics panel, we saw a need to educate our community on the importance of following virus prevention methods. And so, we tried to spread awareness by teaching TAS and local school students (Figure 2). We integrated synthetic biology concepts to explain the difficulty in producing vaccines and why COVID-19 prevention is essential.
Figure 2: Team members taught the concept of virus to a class of eager 1st graders
We also realized the importance of addressing societal issues regarding test kits through our interviews with local doctors and Ms. Athena Hollins, who is a Minnesota state representative and expert in public healthcare (Figure 3). For instance, in the US, Ms. Hollins identified a severe lack of social incentives for impoverished Americans to get tested, so she suggested reimbursement to such populations to encourage testing.
Figure 3: Team members interviewed Ms. Athena Hollins (bottom right panel) and Mr. Rick Brundage (bottom left panel)
In order to gain a better understanding of what the public thinks regarding test kits in general, the team decided to conduct a public survey around Taipei (Figure 4). From the survey results, more than 75% of the respondents confirmed that they would be willing to use a home test kit, which greatly motivated us to bring our product into reality (Figure 5).
Figure 4: Team members asked a pedestrian to help complete our survey
Figure 5: Our survey result shows that more than 75% of our respondents express that they would use a home detection kit.
What would our project look like for an at home test kit? (Figure 1)
Figure 1: Flowchart of our Viral Spiral Project for Viral Diagnosis
Through our modeling hardware piece, we designed a 2 in 1 saliva collection and nucleic acid isolation device. Adding this treated sample into our RCA reaction, a color change will indicate a positive or negative test. Inputting a recorded video of the reaction in our software, we can determine the pH change in the reaction, based on the hue. Our model can then predict the RNA concentration and the corresponding viral load of the individual through the rate of pH change. From a Human Practices perspective, this presents ways to seek immediate medical attention or quarantine procedures for the benefit of oneself and the others around the individual.
We’ve even developed a marketing plan in an attempt to make our test kit available to the public. We’ve included an analysis of our Viral Spiral test kit to other types of virus detection kits on the market and talked to marketing experts to price our kit at a competitive price point.
TAS Science Department
Dr. Sharon Hennessy: Former TAS Head of School
Dr. Grace Cheng Dodge: TAS Interim Head of School
Mr. David Iverson, Science Department Chair
Ms. Joanna Lin, Science Department Assistant
Mr. Andrew Lowman, TAS Upper School Principal
Parents and friends of the TAS_Taipei 2020 Team
Ms. Pearl Fong, Dr. Michael Jewett, Dr. Sui-Yuan Chang, Dr. Wen-Chien Chou, Professor Fore-Lien Huang, Professor Min-Chi Chen, Dr. Chen-Jieun Jan, Dr. Si-Buo Wong, Dr. Kuo-Chin Huang, Dr. Shao-Lun Ho, Ms. Athena Hollins, Mr. Rick Brundage, Ms. Susan Lin International Community Radio Taipei, China Post
