key: cord-0939694-7zrzximg
authors: Aoki, Mateus Nóbrega; de Oliveira Coelho, Bruna; Góes, Luiz Gustavo Bentim; Minoprio, Paola; Durigon, Edison Luiz; Morello, Luis Gustavo; Marchini, Fabricio Klerynton; Riediger, Irina Natassja; do Carmo Debur, Maria; Nakaya, Helder I.; Blanes, Lucas
title: Colorimetric RT-LAMP SARS-CoV-2 diagnostic sensitivity relies on color interpretation and viral load
date: 2021-04-27
journal: Sci Rep
DOI: 10.1038/s41598-021-88506-y
sha: a4683b343a969b3028b3bcb617a80a0db34f4a08
doc_id: 939694
cord_uid: 7zrzximg

The use of RT-LAMP (reverse transcriptase—loop mediated isothermal amplification) has been considered as a promising point-of-care method to diagnose COVID-19. In this manuscript we show that the RT-LAMP reaction has a sensitivity of only 200 RNA virus copies, with a color change from pink to yellow occurring in 100% of the 62 clinical samples tested positive by RT-qPCR. We also demonstrated that this reaction is 100% specific for SARS-CoV-2 after testing 57 clinical samples infected with dozens of different respiratory viruses and 74 individuals without any viral infection. Although the majority of manuscripts recently published using this technique describe only the presence of two-color states (pink = negative and yellow = positive), we verified by naked-eye and absorbance measurements that there is an evident third color cluster (orange), in general related to positive samples with low viral loads, but which cannot be defined as positive or negative by the naked eye. Orange colors should be repeated or tested by RT-qPCR to avoid a false diagnostic. RT-LAMP is therefore very reliable for samples with a RT-qPCR Ct < 30 being as sensitive and specific as a RT-qPCR test. All reactions were performed in 30 min at 65 °C. The use of reaction time longer than 30 min is also not recommended since nonspecific amplifications may cause false positives.

| (2021) 11:9026 | https://doi.org/10.1038/s41598-021-88506-y www.nature.com/scientificreports/ human RNA as negative control) and we found that the method is as specific as RT-qPCR. We also found that the method is specific after testing it with dozens of other respiratory-related viruses. Further, we discuss the presence of a third color cluster never described before that could lead to a wrong diagnosis.

LAMP primers screening. We selected three manuscripts that presented promising results 13, 14, 18 ; these sets of primers were evaluated through the viral strain found in Brazil. As described in "Methods", supernatant RNA extracted from SARS-CoV-2 cultured in Vero cells was quantified by four points in a E-gene standard curve from 50,000 to 50 copies per reaction. We could access a standard curve with amplification efficiency of 98.17% and r 2 of 0.99, with mean Ct ranging from 24.46 (士0.39) and 34.55 (士0. 16 ). Next, the evaluation of different primers' set was made to test RT-LAMP analytical sensitivity for SARS-CoV-2 detection. Figure 1 shows the results obtained using four different primers sets in the presence of the quantified virus RNA (2400 copies) or a non-template control (NTC). A tenfold virus RNA template dilution (240 copies) was also performed (data not shown), for primers sets screening accessed by 2 SARS-CoV-2 RNA concentrations. Our chosen criteria to decide which primer set to proceed, were the color intensities in the reaction tubes and in the agarose gel. According to Fig. 1 , it is possible to notice that the set three 13 was slightly more efficient than the other sets due to its color visualization and agarose gel intensity. Set one 18 and set four 13 presented a weaker gel visualization when compared with the other sets. After the RT-LAMP reactions it was observed that set three 13 had a better color contrast between positive and negative samples when compared to Set two 14 .

Even though it is believed that LAMP primers are easily designed by programs like PrimerExplorer, primerdimers' interactions are recurrent 19 . The inner primers (FIP and BIP) are prone to form hairpin structures, and as a consequence the primers fold back on themselves; thus, the high number of primers (usually six) increases the chances of primer-dimer formation 19 . Therefore, it is important to increase the stress on the system in order to evaluate the occurrence of these interactions, preventing the false-positive results. To assess a more precise analytical sensitivity of the reaction (primers set 3 13 ), a tenfold SARS-CoV-2 RNA dilution from 2.400 to 2,4 viral genome copies per reaction was made. Figure 2 shows both 2.400 and 240 copies with a yellow color whereas 24 and 2,4 remained pink, as observed in negative control. This performance was also confirmed by visualization in 2% agarose gel stained with ethidium bromide. This experiment shows that the analytical sensitivity was between 240 and 24 copies, which represent an amount clinically significant and relevant for this method.

Method sensitivity. To achieve a more precise analytical sensitivity level, we further diluted the RNA using 10 different vials containing 200, 150, and 100 copies per reaction. In the samples with 200 copies, all 10 of the vials turned yellow; consequently, 100% were positive. Using 150 and 100 copies per reaction, 90% and 70%, respectively, changed color to yellow. When the amplification products were observed in agarose gel (Fig. 3) , www.nature.com/scientificreports/ 100% amplification was seen in those containing 150 copies, and 90% amplification was seen when 100 copies were presented. With this, we can observe that with 100 copies presented the colorimetric RT-LAMP reaction is positive by agarose gel visualization in mostly reactions but with no color change, probably due to reduced amplicon formation and consequently minimal pH reduction. This demonstrates that RT-LAMP is a sensitive method for SARS-CoV-2 identification; however, for colorimetric diagnosis this sensitivity is lower, representing a methodological weakness. For evaluation of colorimetric RT-LAMP diagnostic sensitivity, we used 62 SARS-CoV-2 clinical samples confirmed by RT-qPCR, with Ct values for the E gene and RdRp gene ranging from 13.38 to 33.00 and from 15.12 to 32.86, respectively. All samples had Ct for internal control RNAse P less than 33 (Fig. 4 ). When these samples were performed in colorimetric RT-LAMP we observed 79% as positive, 14.5% as indeterminate, and 6.5% negative (Fig. 5 ). All positive and indeterminate samples, independent of colorimetric RT-LAMP result, had amplification products when visualized in agarose gel, demonstrating that color change may not occur despite amplification. It is important to point out that false-negative results in colorimetric RT-LAMP were obtained in 6.5% of samples but again, with amplification visualized in 2% agarose gel, this result was probably due to lower amplicon formation and consequent low pH reduction. Method specificity. Unlike in other reports that used a majority of COVID-19-related samples to evaluate the specificity of the test, we explored a wide range of respiratory viruses that could lead to false-positive results. In this direction, we increased the stress on the test using 131 non-SARS-CoV-2 samples that included 74 healthy subjects and 57 samples with 11 other virus infections, including other coronaviruses. All these samples were collected and diagnosed by the Paraná Central Laboratory (LACEN) in the years 2018 and 2019, i.e., before the SARS-CoV-2 spread in Brazil. Nevertheless, to confirm that these samples were not infected with SARS-CoV-2, all of them were tested using RT-qPCR as described, with no SARS-CoV-2 detection by E-gene or RdRp, and internal control detected in all samples with Ct less than 32.53 (Table 1) . When these certified non-SARS-CoV-2 samples were tested with the colorimetric RT-LAMP, no positive reaction was observed either by color change or by agarose gel stained with ethidium bromide, indicating the high specificity of the method (Fig. 6 ) and representing 100% of diagnostic specificity.

Colorimetric quantification. In order to understand the lower diagnostic sensitivity (79%) obtained by us in comparison with that of other authors, we decided to have an objective color determination by calculating the differences between the absorbance values at 434 nm and 560 nm. Through this spectrophotometric analysis we observed three separated clusters: positive, indeterminate and negative (Fig. 7) , not just two as described in the overall literature. All SARS-CoV-2 colorimetric RT-LAMP scientific reports up to the present time consider only positive and negative results when assessing the diagnostic sensitivity; this interpretation may erroneously increase the real sensitivity. If our samples clustered as indeterminate were designated as positive, our diagnostic sensitivity level would reach 93.5% (74% from positive samples + 14.5% from indeterminate samples), a result consistent with other authors. However, it is important to mention that the colorimetric RT-LAMP reaction is designed to be read by the naked eye: indeterminate samples should not be considered in this context.

According to Thi et al. 12 results, SARS-CoV-2 positive samples with RT-qPCR Ct > 30 maintained the pink color pattern or took more than 30 min to start changing color, determining an affordable limit for colorimetric RT-LAMP. In our study, some SARS-CoV-2 positive samples with RT-qPCR Ct > 30 presented neither yellow nor pink, but an orange color, here clustered as indeterminate. These results could be interpreted as false negatives or false positives by the naked-eye analysis. Therefore, this group of samples is subject to wrong diagnostic treatment or to no treatment at all, thus spreading the virus to others. A colorimetric RT-LAMP as a point-of-care test 20 tested saliva for RT-LAMP with no RNA extraction and reached more than 90% of sensitivity. This is an interesting result since it indicates that the use of saliva associated with RT-LAMP is also possible for the RT-LAMP method. Fowler et al. 21 demonstrated an overall RT-LAMP sensitivity of 97%, but with a 33 Ct cut-off the sensitivity increased to 100%. They also evaluated the RT-LAMP method without RNA extraction, with dilutions from saliva and directly from the swab, and achieved a sensitivity of 67%. Despite we demonstrated a lower colorimetric RT-LAMP positive rate (79%) in comparison with other results, for instance those obtained by Wei et al. 22 (100% with crude samples), L'Helgouach et al. 23 (95,7% with saliva samples), and Haq et al. 24 (100% with extracted samples), with Ct values in RT-qPCR less than 30 we achieved 100% specificity with non-SARS-CoV-2 samples, including related virus infections. Few papers explored diagnostic specificity with other viral infections and with as few clinical samples as we did. Nawattanapaiboon et al. 25 , for example, tested colorimetric RT-LAMP for SARS-CoV-2 method with 13 viruses extracted from clinical samples 27 accessed a large number of other viral infections, looking for 143 other viruses in RT-LAMP for 90 minutes and showed 100% of specificity. Diagnostic specificity is one of the key parameters to validate a diagnostic assay and is necessary because false positive results can lead to different treatments. When LAMP primers are well designed, considering specific and conserved regions, they tend to be specific to the chosen target 28, 29 . Many studies design LAMP primers for the N gene, due to its conserved region. Rohaim et al. 30 , however, designed primers for the RdRp gene and obtained no cross-reactivity with other coronaviruses or four other respiratory viruses. Our results show the high specificity of the colorimetric RT-LAMP test for COVID-19 even when it is exposed to 11 species of other respiratory related viruses. This high diagnostic specificity of the method shows that colorimetric RT-LAMP can be an efficient and reliable tool to SARS-CoV-2 diagnostic during the early stages of the disease, which is essential to control the pandemic. Although RT-qPCR is the gold-standard method to detect COVID-19, it has become evident that this technology is too laborious to effectively stop the virus from spreading. On the other hand, the use of lateral flow tests used to scan the populations has presented high rates of false positives and negatives, becoming a cause of concern instead of a solution. In this context, new technologies need to be introduced to meet the testing demand. The colorimetric RT-LAMP method, described by a few authors as being able to detect SARS-CoV-2, has been evidenced to combine simplicity and accuracy, thereby presenting a viable alternative to fight this disease. Our results show that targeting the N gene with the colorimetric RT-LAMP is a sensitive and specific approach to detect SARS-CoV-2 if the reaction runs at 65°C for 30 minutes. We also proved that the method is robust, stressing the test with 11 respiratory-related viruses. We also report here that the color changes depend on the viral load on the sample. These indeterminate samples represent a delicate cluster that should be carefully analyzed to return a confident result. Nonscientific paper report SARS-CoV-2 colorimetric RT-LAMP results in three clusters described here as positive, indeterminate, and negative. We demonstrated that the color change is not evident and easily distinguished by the naked-eye in an indeterminate cluster, correlating with viral load accessed by RT-qPCR. Corroborating this data, Thi et al. 12 demonstrated that SARS-CoV-2 detection by colorimetric RT-LAMP is directly dependent on viral load and showed that positive samples with a RT-qPCR Ct<30 changed color within the first 30 minutes of reaction. Samples with RT-qPCR Ct > 30 either did not change color or did so at time points >35 min, simultaneously with a color change observed in some of the negative samples. But increasing the time is not helpful since nonspecific amplification reactions may occur. A robust color change in the colorimetric RT-LAMP test for samples with RT-qPCR Ct values between 30 and 35 was observed for only one of 10 samples. This finding suggests a limit for colorimetric RT-LAMP assay corresponding to a Ct ≈ 30. Nawattanapaiboon et al. 25 using RT-qPCR, determined 47 SARS-CoV-2 positive samples with Ct values ranging from 17 to 38, with sensitivity of 95.74% and specificity of 99.95%. A colorimetric RT-LAMP false-negative sample exhibited high RT-qPCR Ct values of 34.17 and 34.93 for ORF1ab-and N-specific primers, respectively. Furthermore, all RT-qPCR SARS-CoV-2 positive samples that yielded Ct values lower than 33 were found by the authors to be true positive in colorimetric RT-LAMP. However, when the Ct values were greater than 33, the percentages of true positive samples decreased to 60% and 40%, for RT-qPCR ORF1ab and N primers, respectively. Again, Schermer et al. 31 using colorimetric RT-LAMP failed to detect the virus in specimens that were positive in diagnostic RT-qPCR at Ct values > 30 for E and S gene. It is important to note that all these experiments www.nature.com/scientificreports/ apparently have been conducted with RT-qPCR validated clinical samples, while colorimetric RT-LAMP as a point-of-care test will be performed alone. Therefore, clinical samples with a lower viral load, here demonstrated by RT-qPCR Ct > 30, may produce false-negative or indeterminate results. More importantly, all these authors ignored the indeterminate sample color that we demonstrated to exist, especially with correlation between RT-qPCR results. For application of colorimetric RT-LAMP for point-of-care purposes, we strongly recommend its use for samples with Ct<30 or in acute cases, as it has been observed in this study the technique detection limit. In short, our results corroborate the data from other authors, showing that colorimetric RT-LAMP can be used as a first-line method for fast and reliable COVID-19 detection. However, color change interpretation should be carefully analyzed since it represents an essential property for the assay performance as a point-of-care tool. Pink color should be considered negative, yellow as positive, and orange as indeterminate. We also verified that the patient viral load is an important limitation on colorimetric RT-LAMP for point-of-care usage, as we demonstrated that RT-qPCR > 30 may return indeterminate or false-negative results. We here suggest that when the color change is not clearly distinguishable, the sample should be repeated or directed to RT-qPCR, and the patient should stay in quarantine meanwhile. We also observed that RT-LAMP will be more effective if used in the first week after the symptoms appear 13, 14, 16 . In short, colorimetric RT-LAMP is a sensitive and specific method that can be used to fight the new coronavirus pandemic if the method limitations are understood and respected. 13, 14, 18 due to the high quality of the data presented. We selected and named four (one to four) primer sets to compare their performance levels (Supplementary Table S1 ). Zhang et al. 13 designed a primer set for the N and ORF region of SARS-CoV-2, and Yu et al. 14 targets the ORF1ab region. Lamb et al. 18 , however, does not specify the target region of their primer set. All the primer were purchased from Integrated DNA Technologies (IDT, United States) and resuspended in at a concentration of 100 μM in nucleasefree water (Invitrogen, United States).

Mix (NEB, England) was used. This kit contains phenol red, a pH indicator that changes color from pink to yellow due to the formation of pyrophosphate ions produced during amplification. Each set of primers contained FIP (Forward Inner Primer) and BIP (Backward Inner Primer) at 1,6 μM each, FOP (Forward Outer Primer) and BOP (Backward Outer Primer) at 0,2 μM each, and FL and BL (Forward Loop and Backward Loop) at 0,4 μM each. The reaction was conducted with a final volume of 25μL, using 12,5μL of WarmStart Colorimetric LAMP 2X Master Mix, 6μL of Primer Oligo, 3μL of RNA, and 3,5μL of RNase Free water. Primers from set one were incubated at 63 °C, while those of sets two, three, and four were incubated at 65 °C, all for 30 min. After this step, all the sets were incubated at 80 °C for 10 min for enzyme inactivation. All the incubations were done using the ProFlex PCR System (Applied Biosystems, United States). Positive reactions presented a yellow color while the negative ones remained pink. Orange samples were considered indeterminate. Human RNA (oncological breast cancer cell line MCF7) was used as negative control (NC). For amplification confirmation, reactions were run on 2% agarose gel electrophoresis (100 V) for 45 min, stained with ethidium bromide and visualized using a UV transilluminator (L-Pix Chemi, Loccus, Brazil). 

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and coronavirus disease-2019 (COVID-19): The epidemic and the challenges

World Health Organization. Coronavirus Disease (COVID-19) Dashboard

Coronaviruses: an overview of their replication and pathogenesis

Evaluation of two automated and three rapid lateral flow immunoassays for the detection of anti-SARS-CoV-2 antibodies

Improved molecular diagnosis of COVID-19 by the novel, highly sensitive and specific COVID-19-RdRp/Hel real-time reverse transcription-pcr assay validated in vitro and with clinical specimens

Development of reverse transcription loop-mediated isothermal amplification assays targeting severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)

COVID-19, an emerging coronavirus infection: current scenario and recent developments -an overview

novel coronavirus disease (COVID-19): paving the road for rapid detection and point-of-care diagnostics

Real-time turbidimetry of LAMP reaction for quantifying template DNA

Loop-mediated isothermal amplification (LAMP) method for rapid detection of Trypanosoma brucei rhodesiense

Sensitive visual detection of AHPND bacteria using loop-mediated isothermal amplification combined with DNA-functionalized gold nanoparticles as probes

A colorimetric RT-LAMP assay and LAMP-sequencing for detecting SARS-CoV-2 RNA in clinical samples

Rapid Molecular Detection of SARS-CoV-2 (COVID-19) Virus RNA Using Colorimetric LAMP

Rapid detection of COVID-19 coronavirus using a reverse transcriptional loop-mediated isothermal amplification (RT-LAMP) diagnostic platform

Rapid and visual detection of 2019 novel coronavirus (SARS-CoV-2) by a reverse transcription loop-mediated isothermal amplification assay

RT-LAMP for rapid diagnosis of coronavirus SARS-CoV-2

Detection of 2019 novel coronavirus (2019-nCoV) by real-time RT-PCR

Rapid detection of novel coronavirus (COVID-19) by reverse transcription-loop-mediated isothermal amplification

Impact of primer dimers and self-amplifying hairpins on reverse transcription loop-mediated isothermal amplification detection of viral RNA

Rapid and extraction-free detection of SARS-CoV-2 from saliva with colorimetric

A highly effective reverse-transcription loop-mediated isothermal amplification (RT-LAMP) assay for the rapid detection of SARS-CoV-2 infection. medRxiv 2020.06.30

Direct diagnostic testing of SARS-CoV-2 without the need for prior RNA extraction

EasyCOV : LAMP based rapid detection of SARS-CoV-2 in saliva

Development Optimization and Validation of RT-LAMP based COVID-19 Facility in Pakistan

Colorimetric reverse transcription loop-mediated isothermal amplification (RT-LAMP) as a visual diagnostic platform for the detection of the emerging coronavirus SARS-CoV-2

Development of a reverse transcription-loop-mediated isothermal amplification as a rapid early-detection method for novel SARS-CoV-2

Symptom screening at illness onset of health care personnel with SARS-CoV-2 infection in king county Washington

Development and clinical application of a rapid and sensitive loop-mediated isothermal amplification test for SARS-CoV-2 infection

Rapid Detection of SARS-CoV-2 Using Reverse transcription RT-LAMP method

Artificial intelligence-assisted loop mediated isothermal amplification (AI-LAMP) for rapid detection of SARS-CoV-2

Rapid SARS-CoV-2 testing in primary material based on a novel multiplex RT-LAMP assay

provided Pasteur funding, RNA from SARS-CoV-2 isolate and revised the manuscript

The authors would like to thank to Fiocruz, Fiotec (Brazilian Foundation for the Scientific 

The authors declare no competing interests.

The online version contains supplementary material available at https:// doi. org/ 10. 1038/ s41598-021-88506-y.Correspondence and requests for materials should be addressed to L.B.Reprints and permissions information is available at www.nature.com/reprints.Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.