In order to create our viral detection test with high sensitivity, high specificity, fast reaction time, room temperature functional capability, and a colorimetric readout, we selected two enzymes for their specific functions: SplintR ligase (PBCV-1 DNA ligase or Chlorella virus DNA ligase) and Φ29 DNA polymerase (Φ29 DNA polymerase). We designed plasmid constructs to maximize the constitutive expression of these two enzymes in Escherichia coli (E. coli), purified these histidine tagged enzymes, and tested it in our Padlock Probe Rolling Circle Amplification detection method.
Team:TAS Taipei/Experiments
Experiments
Experimental Summary
Padlock Probe Rolling Circle Amplification
Theory
For our Viral Spiral viral detection kit, we decided to employ an isothermal methodology known as padlock probe (PLP) rolling circle amplification (RCA). The two ends of the PLP are designed to bind to a specific complementary target sequence. Upon hybridization, a ligase, such as T4 ligase or SplintR ligase, can ligate the ends of the PLP due to a 5’ phosphorylation on the PLP and the presence of ATP as a cofactor (Hamidi & Ghourchian, 2015). The result is a circularized PLP, which can now act as a template for DNA amplification (Hamidi & Perreault, 2019) (Figure 1). If the PLP is not complementary to the target sequence, ligation does not occur (Ge et al., 2013). Without a circularized PLP, subsequent DNA amplification cannot occur, and thus the readout will show a negative result.
Figure 1: The rolling circle amplification RCA process. Upon successful hybridization of the padlock probe to the target, a ligase will circularize the PLP. Then, Φ29 DNA polymerase will catalyze the polymerization of nucleotides to the PLP. Since Φ29 DNA polymerase has helicase activity, once the polymerase reaches the target strand, it will displace the previous strand and continue amplification around the PLP.
The subsequent methods of DNA amplification varies. Some involve the use of the 3’ to 5’ exonuclease activity of Φ29 DNA polymerase to cut the target sequence until the ligation junction is reached, which enables DNA synthesis to proceed (Li et al., 2017). Others use an exonuclease or RNase to achieve the same function or direct annealing of a forward primer to the circularized backbone (RNase H-Assisted RNA-Primed Rolling Circle Amplification for Targeted RNA Sequence Detection | Scientific Reports, n.d., p.). Whatever the case, the DNA polymerase of choice, usually Bst DNA polymerase or Φ29 DNA polymerase, has an existing DNA strand to synthesize DNA strands in circles around the PLP through its helicase activity (Joffroy et al., 2018). A second reverse primer that binds to the newly synthesized DNA strands is sometimes used to further increase DNA synthesis by introducing a branching-like amplification (Hamidi & Ghourchian, 2015; Hamidi & Perreault, 2019; Larsson et al., 2004). Finally, the readout usually uses fluorescent probes that recognize the synthesized strand and is thus quantifiable (Hamidi & Ghourchian, 2015; Larsson et al., 2004, 2010).
Modifications
The primary modification we made to the padlock probe RCA method involves a colorimetric readout. We made use of the fact that for every nucleotide incorporated in DNA amplification, there is a release of acidic hydrogen ions and pyrophosphate (Figure 1). Thus, with a pH indicator we can visualize a color change readout through a change in hydrogen ion concentration. But this also requires removal of all buffers in the solutions of the reaction that might neutralize the hydrogen ions produced. Thus, our reactions were carried out with customized solutions containing no buffers such as Tris-HCl.
Another modification is our use of SplintR ligase (PBCV-1 DNA Ligase and Chlorella virus DNA Ligase) to allow the ligation of our DNA padlock probe and viral RNA target. This enables high specificity through direct RNA targeting. We also used Φ29 DNA polymerase along with SplintR ligase, as both enzymes operate at optimal settings near room temperature and does in fact allow a room temperature reaction as experimentally proven.
To enable the detachment of the target sequence to the circularized padlock probe in a simpler method, we designed universal primers that would bind to the backbone (non-target region) of the padlock probe. As a result, the helicase activity of the polymerase itself would remove the target sequence and allow for amplification to continue. As reports have come up that the exonuclease activity of Φ29 DNA polymerase can involve not only single stranded sequences but double stranded complementary sequences, we added phosphorothioate bonds to disable Φ29 DNA polymerase from digesting it (Li et al., 2017).
All of these traits work together to form our simple, fast, and accurate universal viral detection platform.
Construct Designs
SplintR Ligase
SplintR ligase catalyzes the ligation of adjacent single-stranded DNA splinted by complementary RNA strands (Figure 2) (Biolabs, n.d.-c). SplintR ligase has been previously shown to be capable of differentiating these ligation junctions to single nucleotide polymorph levels and ligate padlock probes (Biolabs, n.d.-c; Krzywkowski & Nilsson, 2017).
Figure 2:Function of SplintR ligase.
SplintR ligase catalyzes the ligation of a 5’-phosphorylated single-stranded DNA end to an adjacent 3’ single-stranded DNA end splinted by complementary RNA strands. The reaction is ATP-dependent.
We obtained the amino acid sequence of the SplintR ligase protein found in Paramecium bursaria Chlorella virus 1. We optimized the DNA sequence for expression in E. coli and removed the PstI cutting site. We attached a 6x histidine tag (6x His-tag) upstream of the SplintR ligase sequence for purification purposes followed by a glycine-serine linker (GS linker) to form our ORF (BBa_K3352001). This served as our open reading frame (ORF; BBa_K3352000) which we flanked with an upstream strong promoter and strong ribosome binding site (RBS) combination (BBa_K880005) and downstream double terminator (BBa_B0015) (Figure 3). This entire composite part was gene synthesized by IDT.
Figure 3: Construct of SplintR ligase with strong promoter, strong RBS and double terminator (BBa_K3352004).
Φ29 DNA polymerase 2
Φ29 DNA polymerase synthesizes new strands of DNA with strand displacement for any existing DNA strands in front (Biolabs, n.d.-b; Li et al., 2017). It also has exonuclease activity from the 3’ to 5’ direction, usually only responsible for the cutting of single stranded nucleic acids but has been shown to be capable of cutting complementary strands as well (Biolabs, n.d.-b; Li et al., 2017) (Figure 4).
Figure 4: Left: Φ29 DNA polymerase (yellow circle) has 3’ to 5’ exonuclease activity and can remove nucleotides from single-stranded DNA. Right: Φ29 DNA polymerase also has DNA polymerase activity, catalyzing the polymerization of deoxynucleotides onto a template.
We obtained the amino acid sequence of Φ29 DNA polymerase from Bacillus phage Φ29. We optimized the DNA sequence for expression in E. coli and removed the EcoRI cutting site to conform to BioBrick assembly standards. Just like our construct design for SplintR ligase, we attached a 6x His-tag upstream of the Φ29 DNA polymerase for purification purposes followed by a GS linker to allow flexibility between tag and Φ29. We then flanked the open reading frame with upstream strong promoter and strong ribosome binding site (RBS) combination (BBa_K880005) and downstream double terminator (BBa_B0015) (Figure 5). This entire composite part was gene synthesized by IDT.
Figure 5: Construct of Φ29 DNA polymerase with strong promoter, strong RBS and double terminator (BBa_K3352005).
Protein Expression and Purification
We transformed our designed plasmids into DH5⍺ E. coli cells. We grew overnight cultures, diluted those cultures, and then grew the cells to log phase. We lysed cells with xTractor Lysis Buffer and purified our His-tagged proteins using Ni sepharose affinity chromatography. In order to check if our proteins were correct, we used SDS-PAGE (XTractorTM Buffer & XTractor Buffer Kit User Manual, n.d.).
Based on our results, our SplintR ligase and Φ29 polymerase constructs that used a strong promoter and strong RBS combination (BBa_K3352004 and BBa_K3352005) did not express an appreciable amount of protein (Figure 6).
Figure 6: We performed Ni sepharose affinity chromatography to purify our His-tagged SplintR ligase (left) and Φ29 DNA polymerase (right). The SDS-PAGE results show that our SplintR ligase and Φ29 DNA polymerase constructs did not express as strongly as we expected.
To verify our SplintR ligase and Φ29 DNA polymerase expression in E.coli, we used their lysate to SDS-PAGE, expecting bands at 35.7 kDa and 68.2 kDa respectively. However, we did not see a strong band at the correct size, which prompted us to redesign our constructs.
Improved Construct Designs
Seeing that purified Φ29 DNA polymerase and SplintR ligase are fundamental to the development of our diagnostic test, we attempted to resolve the issue of low protein expression by replacing the strong promoter in our constructs with a T7 promoter and expressing our protein in BL21(DE3) E. coli. BL21(DE3) strains contain the chromosomal gene T7 RNA polymerase, which is regulated by a lac promoter (Biolabs, n.d.-a). T7 RNA polymerase has been found to be highly selective and efficient in transcribing only the T7 promoter (Arnaud-Barbe, 1998; Biolabs, n.d.-a). Resulting in almost a five-fold faster elongation rate than E. coli RNA polymerase, T7 would be a much stronger promoter of choice . Thus, by using IPTG during protein expression to activate the lac promoter, and thus the T7 RNA polymerase, of our BL21(DE3) E. coli culture, we would effectively significantly increase the production of our enzymes positioned downstream of our T7 promoter (Biolabs, n.d.-a; T7 Promoter System Vectors for Highest Expression Levels in Bacteria, n.d.). We obtained the sequence of the T7 promoter (BBa_J65997) from the Parts Registry and used it to replace the strong promoters on our SplintR ligase and Φ29 DNA polymerase constructs. These composite parts were synthesized by Twist Biosciences and IDT (Figure 7-8).
Figure 7: Our SplintR ligase construct was redesigned to replace the original strong promoter with a T7 promoter (BBa_J65997) to yield our composite part BBa_K3352006.
Figure 8: Our Φ29 DNA polymerase construct was redesigned to replace the original strong promoter with a T7 promoter (BBa_J65997) to yield our composite part BBa_K3352007.
Protein Expression and Purification
We transformed our newly designed plasmids into BL21(DE3) E. coli cells. We grew overnight cultures, diluted the cultures, and grew the cells to OD600 0.5. We then induced expression with 0.1 M IPTG and allowed cultures to grow an additional 2 hours. We harvested cells after the 2 hours and then lysed them with xTractor Lysis Buffer (XTractorTM Buffer & XTractor Buffer Kit User Manual, n.d.)). We purified our His-tagged proteins using Ni sepharose affinity chromatography. In order to check if our proteins were correct, we used SDS-PAGE. Our results show SplintR ligase and Φ29 DNA polymerase migrating at the expected sizes of 35.7 kDa and 68.2 kDa, respectively (Figure 9).
Figure 9a: SDS-PAGE results show that E.coli is able to produce SplintR ligase(BBa_K3352006) and Φ29 DNA polymerase(BBa_K3352007). Bacterial cultures were grown overnight at 37°C, lysed, purified, and prepped for SDS-PAGE. The expected sizes are listed on the side. However, Φ29 DNA polymerase did not express as strongly as we expected, so another trial was done.
Figure 9b: SDS-PAGE results show that the Φ29 DNA polymerase (BBa_K3352007) expressed successfully. We grew bacterial cultures overnight at 37°C, diluted to an OD600 of 0.2, grew them to an OD600 of 0.5, and then collected a 1 mL sample. We added IPTG, grew the cultures for another 4 hours, and then collected another 1mL sample. We pelleted, resuspended, and boiled samples in 1x Sample Buffer to load on SDS-PAGE. The sample with the IPTG expressed the protein more strongly, which suggests that our protein was present.
We also aimed to improve this construct by using pET11a and pET3a vectors with appropriate BioBrick prefixes and suffixes that fulfill the assembly standard. The pET vectors include the T7 promoter, which promotes high level transcription. By utilizing both a T7 promoter, T7 terminator, and an extended UTR sequence around the RBS and before the terminator, we would maximize the protein expression for our enzymes. These composite parts were synthesized by GenScript (Figure 10-11).
Figure 10: Construct of pET T7 Promoter with Modified UTR, Extended RBS, Φ29 DNA polymerase and T7 Terminator (BBa_K3352009)
Figure 11: Construct of pET T7 Promoter with Modified UTR, Extended RBS, SplintR Ligase and T7 Terminator (BBa_K3352008)
Figure 12a: SDS-PAGE results show protein content at different steps of protein purification. A band around 35kDa was not present in the flow through lane (red) or the wash buffer lanes, which corresponds with our expected His-tagged SplintR.
Figure 12b: SDS-PAGE results show protein content at different steps of protein purification. A band around 68kDa was not present in the flow through lane (red) or the wash buffer lanes, which corresponds with our expected His-tagged Φ29.
Our SDS-PAGE results show that both purified proteins migrate at the expected sizes (Figure 12a & Figure 12b). However, due to the presence of unfavorable buffer conditions in our eluted proteins, we performed a buffer exchange to reach the desired pH and storage conditions for our enzymes. We used these proteins for our viral detection test.
Padlock Probe Design
For our padlock probe design, we decided to target the Spike (S) glycoprotein gene for SARS-CoV-2 (C-19) and the Hemagglutinin (HA) gene for Influenza A (INFA) and B (INFB) due to their high copy number in viruses. In order to create a highly specific target sequence, the sequence had to be unique to each virus type while also being able to recognize subtle mutations over time. To that end, we aligned the nucleotide sequences of the S gene from several different SARS-CoV-2 variant strains as well as other related coronaviruses (such as SARS and MERS). We identified a 36bp sequence that was highly conserved between various mutations within the SARS-CoV-2 strains, while also exhibiting significant differences with SARS and MERS (Figure 13). We performed a similar alignment analysis of the HA gene of Influenza A and Influenza B to identify highly specific target sequences for these viruses (Figure 14-15). All sequences were obtained from NCBI GenBank and compiled through BioEdit. Both DNA and RNA synthetic viral targets were synthesized via IDT.
Figure 13: We chose the SARS-CoV-2 target sequence located in the Spike (S) gene as a target based on alignment data of various strains of SARS-CoV-2 (2019nCov), SARS, and MERS. We chose a 36 nucleotide sequence that is perfectly conserved with those of SARS-CoV2 while showing minimal matching with SARS and MERS.
Figure 14: We chose the Influenza A target sequence located in the hemagglutinin gene as a target based on alignment data of various strains of Influenza A (H1N1pdm09) and Influenza B (Victoria Lineage). We chose a 36 nucleotide sequence that is perfectly conserved with those of H1N1 Influenza A while not matching those of Influenza B.
Figure 15: We chose the influenza B target sequence located in the hemagglutinin gene as a target based on alignment data of various strains of influenza A (H1N1pdm09) and influenza B (Victoria Lineage). We chose a 36 nucleotide sequence that is perfectly conserved with those of H1N1 influenza A while not matching those of influenza B.
To target these viruses, we designed PLP sequences by adding half of the reverse complement of the target sequence to each side of a standard and tested PLP backbone. For ligation to occur, a phosphate group modification was added to the 5’ end of the probe. Our primers that help amplify the detection signal also incorporate the phosphorothioate modification on the 3’ end to prevent degradation against the exonuclease activity of Φ29 DNA polymerase.
Preliminary RCA Test Using DNA Targets
To ensure that our test functions in vitro, we first tested our RCA technique with synthetic DNA targets of the viruses we selected. In other words, these are DNA forms of RNA gene fragments from Influenza A (H1N1pdm09), Influenza B (Victoria Lineage), and SARS-CoV-2, which act as a proxy for viral RNA.
For our initial RCA test using DNA, we made use of T4 DNA ligase that is commercially available, to ligate nicks in duplex DNA, instead of SplintR ligase. Although the optimal temperature for T4 ligase is 16°C, our results show that it works just as fine at room temperature.
Figure 16a: The RCA reaction for DNA Target. T7 ligase is used.
Figure 16b: The RCA reaction for RNA Target. SplintR ligase is used in addition to the RNase Inhibitor.
After optimizing our test, we found that we were able to add all RCA components into one tube for the room temperature reaction to take place. We would add the synthetic viral DNA target, the target specific padlock probe, T4 ligase, deionized water, NaOH, and a ligation solution (DTT, MgCl2, ATP) which serves the hybridization and ligation part of the test. The forward primer, reverse primer, phenol red, Φ29 DNA polymerase, an amplification solution (DTT, MgCl2, (NH4)2SO4, dNTPs, NaOH), and deionized water for the DNA and signal amplification to occur for a colorimetric readout (Figure 16). Of note, no Tris or other buffers were added to the reaction.
Confirmation of Nucleic Acid Amplification
During the amplification process, the incorporation of dNTPs into the sequence synthesized results in the formation of PPi and H+. As more and more H+ ions are made, the pH indicator changes from purple to yellow due to a drop in pH. Although we know of no other factors that could cause the same color change, we wanted to validate that our RCA test was functioning and changing colors due to DNA amplification.
To confirm that DNA amplification was occuring, we performed RCA reactions where a positive target was added and no target was added. We removed an aliquot of the reaction 30 min and 40 min into the reaction. Upon removal of each aliquot, we immediately heat shocked at 80C to inactivate the enzymes and stop the RCA reaction. We then ran the samples on an agarose gel to visualize any DNA.
Figure 17: Comparison between viral load of SARS-CoV-2 after 30 minutes of amplification and 40 minutes of amplification. These samples were loaded with 6X concentrated DNA loading dye (geneaid) into 1% agarose gel.
Our results show a significant amount of DNA remaining in the wells in the samples where we added a positive target (Figure 17, + vs -). In addition, these samples exhibited significant smearing of DNA across the entire lane of the gel. These results strongly suggest that DNA amplification occurred only in the samples where a positive target was added. Furthermore, the amplification resulted in highly branched and long nucleotide sequences, which prevented the samples from migrating through the gel. The smearing is likely due to a mixed population of varying nucleotide sequences. Lastly, we observed that the amount of DNA increased in the 40 minute time point compared to the 30 minute time point, which further supports our model that DNA amplification is occurring rapidly over time in our RCA reaction. These results demonstrate that the color change was in fact due to H+ concentration change that occurred via DNA amplification and not due to other factors unrelated to DNA synthesis.
RCA Test Using RNA Targets
After validating the general theory of our test, we moved onto RNA testing which meant adding the same components for the hybridization and ligation process except we replaced T4 DNA ligase with SplintR ligase and we added Protector RNase Inhibitor (Sigma) to prevent RNA degradation.
Video 1: RCA reaction ran using INFA DNA: In tube G, a reaction was run with the INFA padlock probe and no target present. In tube H, a reaction was run with an INFA padlock probe and 0.0625uM INFA DNA target.
Video 2: RCA reaction ran using C19 RNA: Since this is a C19 test, C19 padlock probes were used in all tubes. In tube A, no target was present. In tube B, a SARS target was present. In tube C, a SARS-CoV-2 target was present.
Table 1: RCA for Synthetic viral DNA targets
Table 2: RCA for Synthetic viral RNA targets
Figure 18a: INFA Test DNA vs RNA, a comparison between using the DNA and RNA Test for Influenza A
Figure 18b: INFB Test DNA vs RNA, a comparison between using the DNA and RNA Test for Influenza B
Figure 18c: C-19 Test DNA vs RNA, a comparison between using the DNA and RNA Test for SARS-CoV-2
Our results show that DNA testing for RCA typically yields faster color change results than that of RNA testing. The pH over time graph for Influenza A, Influenza B, and COVID-19, according to Figure 18a, 18b and 18c respectively, show that the DNA and the RNA test have similar trend lines. Although the pH of the DNA test (shown in orange) decreases faster than the RNA test (shown in blue) for some, our results show that both the DNA and RNA test decreases rapidly, which means our RCA test is still functional with RNA targets. Testing both DNA and RNA viruses are crucial because although the high impact viruses we selected to detect, Influenza A (H1N1pdm09), Influenza B (Victoria Lineage), SARS-CoV-2, are RNA viruses, other notable viruses such as smallpox and papillomaviruses are DNA viruses (Sanjuán et al., 2016).
Analytical Sensitivity Test
Another test we conducted concerns the analytical sensitivity of our assay. In these tests, we selected the SARS CoV-2 synthetic viral target and proceeded to conduct serial dilutions to test RCA under different concentrations. Using a micromolar concentration as a standard, we then compared the time of color change that occurred to samples at nanomolar, picomolar, femtomolar, and attomolar concentrations (Table 3).
The table below indicates the number of RNA strands that correlates with each concentration.
Table 3: Conversion of RNA Concentration to Number of RNA strands.
Video 3: Tube A served as a control where we ran the reaction with the SARS-CoV-2 padlock probe but without any target. Tubes B to I have concentrations of 0.0625uM, 0.00625uM, 0.0625nM, 0.027pM, 0.0027pM, 0.00027pM, 0.027fM and 0.027aM of SARS-CoV-2 target respectively.
Figure 19: Sensitivity Test on C-19 target. Testing the sensitivity of the test using different concentrations of C-19
Our results demonstrate that a color change occurs in all concentrations until the attomolar concentration is reached. Though some of the lower concentrations relied on a longer reaction time, it shows that our assay is highly sensitive in detecting synthetic RNA strands (Figure 19). Specifically, we can still detect up to 16.26 synthetic strands while differentiating from a negative test.
Specificity Test
In addition to determining how sensitive our test can be, we also wanted to determine the specificity of our assay. This meant that we had to determine whether differing sequences of synthetic viral targets that reflect different virus strains can be differentiated by our test. To find out, we used no target, a mismatched target, and a matched target for each of our tests as determined by our padlock probe. Each padlock probe has a specific sequence that is only fully complementary to its corresponding target, which should only instigate a reaction (Table 4).
Table 4: Different Tests Ran to Determine the Specificity of Our PLP
The table below shows the different synthetic RNA target and padlock probe combinations used in each reaction to test the specificity of our assay. (+) means the presence of the target while (-) means a lack thereof.
The SARS-CoV-2 and the SARS targets which are both coronaviruses, differ by 18 nucleotides in alignment. The Influenza A and Influenza B targets which are both influenza viruses, differ by 27 nucleotides in alignment.
Video 4: Specificity test using SARS-CoV-2 RNA and Padlock probe:
All tubes contain the SARS-CoV-2 padlock probe. Tube A has no RNA target. Tube B contains the MERS RNA target. There will be no color change in Tube B since the sequence of the target does not match the sequence of the padlock probe. Tube C contains the SARS-CoV-2 RNA target. The reaction will occur because the sequence for the padlock probe matches the complementary sequence of the target.
Video 5: Specificity test using INFA and INFB RNA and their corresponding Padlock probe: Reactions in tubes A, B, C contains the INFA padlock probe and reactions in tubes D, E, F contains the INFB padlock probe. Tubes A and D have no RNA target. Tubes B and E contain the SARS-CoV-2 RNA target where the sequence of the target does not match the sequence of the padlock probe. Tubes C and F contain the correct INFA and INFB RNA padlock probe with the correct RNA target. The reaction will occur because the sequence for padlock matches the complementary sequence of the target.
Figure 20: C-19 Specificity Test
Our results show that when the C-19 padlock probe is used, the pH of the C-19 target solution has the greatest change. As expected, the solution with no target has little to no pH change. The SARS target has a slight pH change due to its similarities with the C-19 nucleotide sequence, but the color difference is still significant enough to distinguish between the two.
Figure 21: INFA Specificity Test
Our results highlight the fact that our padlock probe could detect the difference between the different targets. Although the color and pH of both targets change, the rate of change for the correct target was much faster than the incorrect target.
Figure 22: INFB Specificity Test
Our results highlight the fact that our padlock probe could detect the difference between the different targets. Although the color and pH of both targets change, the rate of change for the correct target was much faster than the incorrect target.
Despite similar nucleotide sequences, a mismatched viral sequence will result in a slower color change than that of when the virus sequence actually tested is present. In our SARS-CoV-2 test, the presence of SARS-CoV-2 target sequence resulted in a notably faster reaction time than in the presence of a SARS target sequence. Similar results were found for the Influenza tests, using the target combinations listed above. The notable time difference in reaction times allows us to differentiate different viruses in the same type, whether that is the influenza virus or the coronavirus.
References
A Covid-19 Nasal Swab Test Punctured Woman’s Brain Lining and Leaked Brain Fluid From Her Nose. (n.d.). Retrieved October 22, 2020, from https://www.news18.com/news/buzz/a-covid-19-nasal-swab-test-punctured-womans-brain-lining-and-leaked-brain-fluid-from-her-nose-2927009.html
Alhassan, A., Li, Z., Poole, C. B., & Carlow, C. K. S. (2015). Expanding the MDx toolbox for filarial diagnosis and surveillance. Trends in Parasitology, 31(8), 391–400. https://doi.org/10.1016/j.pt.2015.04.006
At-Home Saliva Tests for COVID-19 Could Be Coming Soon. (n.d.). Time. Retrieved October 23, 2020, from https://time.com/5891887/covid-19-saliva-spit-test/
Bosman, J., Stockman, F., & Fuller, T. (2020, February 18). ‘Are You Sick?’ For Asian-Americans, a Sneeze Brings Suspicion. The New York Times. https://www.nytimes.com/2020/02/16/us/coronavirus-american-mood.html
CDC. (2020, February 11). Overview of Testing for SARS-CoV-2 (COVID-19). Centers for Disease Control and Prevention. https://www.cdc.gov/coronavirus/2019-ncov/hcp/testing-overview.html
Commissioner, O. of the. (2019, July 9). Emergency Use Authorization of Medical Products and Related Authorities. U.S. Food and Drug Administration; FDA. https://www.fda.gov/regulatory-information/search-fda-guidance-documents/emergency-use-authorization-medical-products-and-related-authorities
Commissioner, O. of the. (2020a). Coronavirus Testing Basics. FDA. https://www.fda.gov/consumers/consumer-updates/coronavirus-testing-basics
Commissioner, O. of the. (2020b). Emergency Use Authorization. FDA. https://www.fda.gov/emergency-preparedness-and-response/mcm-legal-regulatory-and-policy-framework/emergency-use-authorization
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
da Costa Lima, M. S., Zorzenon, D. C. R., Dorval, M. E. C., Pontes, E. R. J. C., Oshiro, E. T., Cunha, R., Andreotti, R., & de Fatima Cepa Matos, M. (2013). Sensitivity of PCR and real-time PCR for the diagnosis of human visceral leishmaniasis using peripheral blood. Asian Pacific Journal of Tropical Disease, 3(1), 10–15. https://doi.org/10.1016/S2222-1808(13)60003-1
Emergency Use Authorization of Medical Products. (n.d.). 49.
False-Negative Rate of RT-PCR SARS-CoV-2 Tests. (n.d.). American College of Cardiology. Retrieved June 5, 2020, from http%3a%2f%2fwww.acc.org%2flatest-in-cardiology%2fjournal-scans%2f2020%2f05%2f18%2f13%2f42%2fvariation-in-false-negative-rate-of-reverse
Frasier, S. L. (n.d.). Coronavirus Antibody Tests Have a Mathematical Pitfall. Scientific American. https://doi.org/10.1038/scientificamerican0720-12
Ge, Y., Zhou, Q., Zhao, K., Chi, Y., Liu, B., Min, X., Shi, Z., Zou, B., & Cui, L. (2017). Detection of influenza viruses by coupling multiplex reverse-transcription loop-mediated isothermal amplification with cascade invasive reaction using nanoparticles as a sensor. International Journal of Nanomedicine, Volume 12, 2645–2656. https://doi.org/10.2147/IJN.S132670
Greenwood, M. (2020, August 15). Yale’s rapid COVID-19 saliva test receives FDA emergency use authorization. YaleNews. https://news.yale.edu/2020/08/15/yales-rapid-covid-19-saliva-test-receives-fda-emergency-use-authorization
How Coronavirus Racism Infected My High School | NYT Opinion. (n.d.). Retrieved October 25, 2020, from https://www.youtube.com/watch?v=f5ccbJcqlUo&ab_channel=TheNewYorkTimes
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
Jamal, A. J., Mozafarihashjin, M., Coomes, E., Powis, J., Li, A. X., Paterson, A., Anceva-Sami, S., Barati, S., Crowl, G., Faheem, A., Farooqi, L., Khan, S., Prost, K., Poutanen, S., Taylor, M., Yip, L., Zhong, X. Z., McGeer, A. J., Mubareka, S., … Walmsley, S. (n.d.). Sensitivity of Nasopharyngeal Swabs and Saliva for the Detection of Severe Acute Respiratory Syndrome Coronavirus 2. Clinical Infectious Diseases. https://doi.org/10.1093/cid/ciaa848
Lalkhen, A. G., & McCluskey, A. (2008). Clinical tests: Sensitivity and specificity. Continuing Education in Anaesthesia Critical Care & Pain, 8(6), 221–223. https://doi.org/10.1093/bjaceaccp/mkn041
Mervosh, S., & Fernandez, M. (2020, August 7). ‘It’s Like Having No Testing’: Coronavirus Test Results Are Still Delayed. The New York Times. https://www.nytimes.com/2020/08/04/us/virus-testing-delays.html
Nasopharyngeal swab vs. Saliva for COVID-19 diagnosis. (2020, May 21). News-Medical.Net. https://www.news-medical.net/news/20200521/Nasopharyngeal-swab-vs-saliva-for-COVID-19-diagnosis.aspx
Policy for Coronavirus Disease-2019 Tests During the Public Health Emergency (Revised)—Immediately in Effect Guidance for Clinical Laboratories, Commercial Manufacturers, and Food and Drug Administration Staff. (n.d.). 20.
Principles of Epidemiology: Lesson 5, Section 2|Self-Study Course SS1978|CDC. (2020, May 11). https://www.cdc.gov/csels/dsepd/ss1978/lesson5/section2.html
Roberts, J., Bebenek, K., & Kunkel, T. (1988). The accuracy of reverse transcriptase from HIV-1. Science, 242(4882), 1171–1173. https://doi.org/10.1126/science.2460925
Skalka, A. M., & Goff, S. (Eds.). (1993). Reverse transcriptase. Cold Spring Harbor Laboratory Press.
Tahamtan, A., & Ardebili, A. (2020). Real-time RT-PCR in COVID-19 detection: Issues affecting the results. Expert Review of Molecular Diagnostics, 20(5), 453–454. https://doi.org/10.1080/14737159.2020.1757437
The Pandemic of Xenophobia and Scapegoating. (n.d.). Time. Retrieved October 25, 2020, from https://time.com/5776279/pandemic-xenophobia-scapegoating/
Tholen, D., Linnet, K., Kondratovich, M., Armbruster, D., Garrett, P., Jones, R., Kroll, M. H., Lequin, R., Pankratz, T., Scassellati, G. A., Schimmel, H., & Tsai, J. (2004). Protocols for Determination of Limits of Detection and Limits of Quantitation; Approved Guidelines. Undefined. /paper/Protocols-for-Determination-of-Limits-of-Detection-Tholen-Linnet/9afc36c690cfbf76f9251464045564a5ee073d7b
Why It Takes So Long To Get Most COVID-19 Test Results. (n.d.). NPR.Org. Retrieved October 22, 2020, from https://www.npr.org/sections/health-shots/2020/03/28/822869504/why-it-takes-so-long-to-get-most-covid-19-test-results
Zhang, S. (2020, March 23). Why a Tiny Colorado County Can Offer COVID-19 Tests to Every Resident. The Atlantic. https://www.theatlantic.com/science/archive/2020/03/coronavirus-tests-everyone-tiny-colorado-county/608590/