A new SYBR Green real-time PCR to detect SARS-CoV-2

Phylogenetic analysis has demonstrated that the etiologic agent of the 2020 pandemic outbreak is a betacoronavirus named SARS-CoV-2. For public health interventions, a diagnostic test with high sensitivity and specificity is required. The gold standard protocol for diagnosis by the Word Health Organization (WHO) is RT-PCR. To detect low viral loads and perform large-scale screening, a low-cost diagnostic test is necessary. Here, we developed a cost-effective test capable of detecting SARS-CoV-2. We validated an auxiliary protocol for molecular diagnosis with the SYBR Green RT-PCR methodology to successfully screen negative cases of SARS-CoV-2. Our results revealed a set of primers with high specificity and no homology with other viruses from the Coronovideae family or human respiratory tract pathogenic viruses, presenting with complementarity only for rhinoviruses/enteroviruses and Legionella spp. Optimization of the annealing temperature and polymerization time led to a high specificity in the PCR products. We have developed a more affordable and swift methodology for negative SARS-CoV-2 screening. This methodology can be applied on a large scale to soften panic and economic burden through guidance for isolation strategies.


Scientific Reports
| (2021) 11:2224 | https://doi.org/10.1038/s41598-021-81245-0 www.nature.com/scientificreports/ To this end, we developed an auxiliary protocol for molecular diagnosis involving RT-PCR with a SYBR Green methodology to detect negative cases of SARS-CoV-2. This protocol will maximize the cost-benefit of viral detection and accelerate availability by the use of conventional kits on a large scale in molecular biology laboratories.

Results
Validation of the primer sequences. The Beacon Designer platform showed that there were no secondary structures, no hairpins, no homodimers, and no cross dimer formation in the primer set sequences. Primer-BLAST analysis of both the hCOVassay1 and hCOVassay2 primer sequences showed that they are present only in the target SARS-CoV-2 genome. When base complementarity of the primer set was searched against the human genome, there was no similarity, reinforcing that the viral genome alone would be amplified. In addition, we verified the primers' ability to anneal in the genomes from other airways circulating opportunistic microorganisms. No relevant homologies were found, i.e., the primers will not generate unspecific amplifications. The searches are shown in Supplementary Table S1 and S2. The primers from hCOVassay1 and hCOVassay2 have proven to be specific only for the SARS-CoV-2 genome.
RNA extraction and reverse transcription optimization. RNA was extracted using a total RNA extraction kit. The results were below 1 ng/μL according to quantification with the NanoDrop fluorometer. The values of the total RNA extracted were on the pg scale, below the equipment's detection threshold.
After reverse transcription, the mean amount of ssDNA obtained was 1394 ng/μL (SD = 26.9) for the hCO-Vassay1 primer, 1327 ng/μL (DP = 107.6) for the hCOVassay2 primer (3′ primer methodology), and 727 ng/μL (SD = 27.3) for the random primers methodology. The ssDNA purity levels were very similar in both techniques. The random primers technique was chosen to produce ssDNA in the first validation stage, since it is easier and quicker than the 3′ primer methodology.
Template DNA concentration curve for quantitative PCR detection. The manufacturer's recommended temperatures, time per cycle, and number of cycles were applied to amplify the targeted DNA. The amplifications with the 3′ primer technique and master mix without UDG activation using the hCOVassay1 primer and DNA dilutions of 500 ng, 100 ng and 50 ng yielded cycle threshold (CT) values of 34, 35 and 37, respectively (Fig. 1A). Amplifications with the hCOVassay2 3′ primer generated CT values of 35, 37 and 37, respectively (Fig. 1B). Amplifications of ssDNA reverse transcribed with random primers and master mix without UDG activation using the hCOVassay1 with dilutions of 100 ng, 50 ng and 10 ng yielded CT values of 32, 33 and 36, respectively (Fig. 1C). When the hCOVassay2 primer was used to amplify the same dilutions of the template DNA, the CT values were 33, 34 and 36, respectively (Fig. 1D). The CT value for the negative controls using 3′ primer ssDNA methodology was 34 for the hCOVassay1 primer set and 36 for the hCOVassay2 primer set (Fig. 1E).
When master mix with UDG activation was used with the 3′ primer technique using 500 ng, 100 ng and 50 ng dilutions of the template DNA, the CT values yielded for the hCOVassay1 primer set were 34, 34 and 37, respectively ( Fig. 2A). The hCOVassay2 primer set generated CT values of 31, 34 and 37, respectively (Fig. 2B). For the reverse transcriptase technique using the random primer methodology and master mix with UDG activation, template DNA with dilutions of 100 ng, 50 ng and 10 ng yielded CT values of 29, 30 and 32, respectively, for the hCOVassay1 primer set (Fig. 2C). When the hCOVassay2 primer set was used, the CT values were 30, 31 and 33, respectively (Fig. 2D). The CT (cut-off) value for the negative controls using 3′ primer ssDNA methodology was 38 for the hCOVassay1 and hCOVassay2 primer set (Fig. 2E).
Thus, the 3′ primer methodology for synthesis of ssDNA was selected as the best option since the parameters of the amplification curves showed less nonspecific signal and a higher level of reliability.
Melt curve analysis. The melt curve of all amplified SARS-CoV-2-positive samples with both sets of primers produced similar dissociation curves patterns and gave a clear and distinct melt peak ( Fig. 3A-D). However, a clearly distinct dissociation curve was obtained for negative control samples that did not align with the melt curve obtained from the positive samples, suggesting that the melt curve peaks generated from the negative samples were due to nonspecific signals or the formation of primer dimers. This parameter is very important in the analysis of specificity of curves for the SyBr Green methodology.
Visualization of RT-PCR amplicons. The RT-PCR amplicons were separated by electrophoresis, and the products were analyzed under UV to confirm negative and positive samples. Figure 4 shows the 102 bp amplicon corresponding to SARS-CoV-2-positive samples, while the last two lanes show negative samples. When master mix with or without UDG activation was used, the negative control samples did not have a 111 bp band for the hCOVassay1 primer set or a 102 bp band for the hCOVassay2 primer set ( Fig. 4: lanes 7 and 8).
Optimization of amplification parameters. The amplification parameters for each primer were modified (primer annealing temperature and time and Taq DNA polymerization time), which improved the capture signal during the RT-PCR assay, mainly when the master mix with UDG activation was used. The parameter alterations reduced possible nonspecific annealing and hindered the formation of double strands larger than 200 bp. Figure 5 shows that the RT-PCR amplification curves obtained from the use of master mix without UDG activation produced a faint SYBR signal for the negative controls for SARS-CoV-2 (Fig. 5A,B), although for both primer sets no amplicons can be visualized after separation by electrophoresis (Fig. 5C). www.nature.com/scientificreports/ The SYBR Green master mix with UDG activation proved to be very efficient and reliable in terms of nonspecific signals. The master mix with UDG activation was more appropriate for avoiding false positives in uninfected human cDNA samples, maintaining the specificity of the amplification signal (Fig. 6).
The Supplementary Table S3 shows the pre-established parameters, as recommended by the manufacturer, and alterations made to improve the detection method are described here.
Quantifiable control and detection curve. It was possible to detect the presence of the virus by observing the amplification curve from 20 copies of SARS-CoV-2 for both primers. Regarding the detection capacity, less than 20 ng of total ssDNA after transcription using the 3′ primer did not generate a considerably safe amplification curve for virus detection. ssDNA equal to or greater than 200 ng were capable of being detected with great reliability. The comparison of the amplification curves showed that 200 ng of ssDNA corresponds to approximately 200 copies of the SARS-CoV-2 virus in the sample (Fig. 7). (E) Amplification curves of negative controls produced using the two primer sets using 3′ primer ssDNA method. The negative control was from a negative patient.  www.nature.com/scientificreports/ primers to generate first-strand DNA, RT-PCR performed with master mix with UDG activation and optimized amplification parameters. Of these 41 samples, 33 were negative and 8 were positive according to this new protocol. All samples were also tested through the CDC protocol for SARS-CoV-2. All 33 negative samples were confirmed, and 7 positive samples were confirmed by the CDC protocol. Only one positive sample according to the SyBr Green protocol was negative in the CDC protocol. The amplification curves of the two set of primers showed a positive signal for confirming positive samples and did not show any signal of amplification curve in negative samples. For confirmation, some samples were submitted for electrophoresis in 2% agarose showing absence of amplicon band. The example of this validation methodology is described in Fig. 8.   This protocol was applied in approximately 3000 samples. We randomized 300 samples tested from April to September 2020. From these analyzed samples, 48 were positive in SyBr Green and TaqMan protocol and 252 samples were negative in both SyBr Green and TaqMan protocol. A large fraction of samples was tested with the TaqMan protocol only when they were positive for SyBr Green protocol. The examples of this analysis (amplification and dissociation curves) are described in Supplementary Figs. S1 and S2.

Discussion
The SYBR Green and TaqMan techniques are routinely used in real-time PCR. Due to its simple design, easy configuration and low cost, the SYBR Green detection methodology is predominantly used for the detection and amplification of nucleic acids. However, the TaqMan methodology uses an additional labeled probe, significantly  www.nature.com/scientificreports/ increasing the sensitivity and specificity of the assay due to the reporter dye's conjugation to the specific oligonucleotide sequence of the probes, which is capable of emitting fluorescence. The TaqMan assay is considered the main method to detect and quantify human pathogens with a low copy number, including viruses 13 . Nonspecific primer binding, which results in the production of unwanted PCR products and the formation of primer dimers, can significantly influence the sensitivity and reliability of the PCR signals. Therefore, the choice of specific primers and their in silico validation, followed by RT-PCR parameter optimization, plays a fundamental role in the success of an RT-PCR assay when SYBR Green is used.
The hCOVassay1 primer set used in this study, given its complementary to the SARS-CoV-2 sequence, had a high specificity index and generated a 111 bp PCR amplicon. This primer has no homology with the genomic sequences of other viruses from the Coronaviridae family or other known human respiratory viruses except rhinoviruses/enteroviruses, which possess a 5 base-pair mismatch with the forward primer and 4 distinct nucleotides on the reverse primer but would generate an amplicon of 409 bp. The forward primer also has five nucleotides homologous to the Legionella spp. genome, and the reverse primer has four mismatches, capable of generating an amplicon of 3591 bp. The hCOVassay2 primer set has four and three nucleotides homologous with the Legionella spp. genome in the forward primer and the reverse primer, respectively. This set of primers can produce amplicons of 3243 bp and 500 bp, respectively.
The primer annealing temperature and TAQ DNA polymerization time were optimized based on information obtained from the Primer Blast (NCBI) data for each sequence, which allowed an increase in amplification specificity for the PCR products. Annealing temperatures of 54 °C were calculated according to the number of CG/AT bases in each primer pair. In addition, the 20-s polymerization time reduced the possibility of unspecific signals generated by partial primer annealing and/or primer dimers.
The standard WHO methodology uses the TaqMan probe technique to detect SARS-CoV-2. This methodology is extremely effective and can even accurately discriminate between SARS-CoV-2 and SARS-CoV infections. The Charité Protocol, developed at the University of Charité, Berlin, uses four different probes to identify SARS-CoV-2. This test is very expensive, which limits its large-scale application. The U.S. Centers for Disease Control and Prevention (CDC) also has a standardized protocol with four different TaqMan probes that can accurately determine SARS-CoV-2 infection. Both the Charité and the CDC protocols involve reagents and probes that are now scarce and expensive on the world market.
The present study established a screening strategy for SARS-CoV-2 through a molecular assay using the SYBR Green methodology. The proposed method harbors a low cost with the potential for large-scale screening of symptomatic and/or asymptomatic individuals. This screening strategy will allow communities or companies to screen for SARS-CoV-2-negative individuals on a large scale. However, when the test shows a positive result, it is advised that the patient undergo the molecular tests established by the WHO or CDC or even a serological test after the period required to detect antibodies, seven to fourteen days. Therefore, the most scarce and expensive tests could be applied only to a fraction of the target population ( Supplementary Fig. S3).
In this study, we established a fast and low-cost method for negative SARS-CoV-2 screening. We suggest that negative by both sets of primers in RT-PCR signal analysis in linear mode can certify the absence of virus. Samples whose signal is compatible with the positive control in one of the two tested primers must be repeated and/or treated as indeterminate. Double-positive samples for the tested primers should be confirmed according to WHO or CDC protocols.
Double-negative patients can resume normality regarding social isolation, while patients who are positive for one or both proposed primers should be immediately tested according to recommended protocols or wait 7 to 14 days for serological tests, also known as rapid tests.
Due to its low cost and processing speed, this methodology can be applied on a large scale, providing peace of mind for those being tested and their peers, as well as guidance for social isolation protocols.
We aimed to develop a low-cost diagnostic test capable of detecting SARS-CoV-2 in the oropharyngeal mucosa of both symptomatic and asymptomatic patients to help reestablish normal work routines as well as to remove SARS-CoV-2-negative individuals from isolation (with appropriate medical follow-up).

Methods
Clinical samples. This study was approved by the Research Ethics Committee of Pontifical Catholic University of Rio Grande do Sul (PUCRS) with approval number 3.977.510. All participants provided written informed consent prior to inclusion in this study. All methods were performed in accordance with relevant guidelines and regulations. To determine the specificity of the primer set, the detection curve and possible cross-reactions, a sample obtained from the Clinical Analyses Laboratory of the Hospital de Clínicas de Porto Alegre, RS, Brazil was used. The sample was collected from an inpatient with suspected COVID-19 who attended the São Lucas da PUCRS hospital. The sample was sent to the Laboratório Central de Saúde Pública do Rio Grande do Sul (LACEN) and Fleury Laboratory for determination of SARS-CoV-2 positivity through the Center of Control Disease and Prevention (CDC) protocol.
In silico study of primer specificity. Two pairs of primers complementary to two different regions of the viral RNA sequence of the isolated coronavirus (Whuan-H1/NC_045512.2) were tested (Supplementary  Table S1). First, the primers were input into Beacon Designer Software (PREMIER BIOSOFT INTERNA-TIONAL) to search their molecular structures based on test type. The SYBR Green option was selected, and the parameters were adjusted accordingly. The primers were then input into the Prime-BLAST platform (NCBI) and evaluated for their specificity with the sequence of the SARS-CoV-2 genetic material. Searches were also performed for similarities within regions of the human genome and of the main respiratory and opportunistic viruses and pathogens to determine whether the primers flanked the regions of interest and amplified only the www.nature.com/scientificreports/ SARS-CoV-2 genetic material. The input SARS-CoV-2 primers were compared with the genomes of the microorganisms shown in Supplementary Tables S1 and S2.

Data availability
The datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request.