Extraction-free protocol combining proteinase K and heat inactivation for detection of SARS-CoV-2 by RT-qPCR

Abstract

Real-time reverse transcription PCR (RT-qPCR) is the gold-standard technique for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) detection in nasopharyngeal swabs specimens. The analysis by RT-qPCR usually requires a previous extraction step to obtain the purified viral RNA. Unfortunately, RNA extraction constitutes a bottleneck for early detection in many countries since it is expensive, time-consuming and depends on the availability of commercial kits. Here, we describe an extraction-free protocol for SARS-CoV-2 detection by RT-qPCR from nasopharyngeal swab clinical samples in saline solution. The method includes a treatment with proteinase K followed by heat inactivation (PK+HID method). We demonstrate that PK+HID improves the RT-qPCR performance in comparison to the heat-inactivation procedure. Moreover, we show that this extraction-free protocol can be combined with a variety of multiplexing RT-qPCR kits. The method combined with a multiplexing detection kit targeting N and ORF1ab viral genes showed a sensitivity of 0.99 and a specificity of 0.99 from the analysis of 106 positive and 106 negative clinical samples. In conclusion, PK+HID is a robust, fast and inexpensive procedure for extraction-free RT-qPCR determinations of SARS-CoV-2. The National Administration of Drugs, Foods and Medical Devices of Argentina has recently authorized the use of this method.

Fig 1. Proteinase K improves the performance of the heat inactivation method in RT-qPCR determinations of SARS-CoV-2.

(a-b) Positive (#8, #11, #12, #16, #19, #20) and negative (#1, #2, #3) nasopharyngeal swab samples were processed by heat inactivation (HID, 98°C for 5 min); proteinase K treatment followed by heat inactivation (PK+HID, 55°C for 15 min and 98°C for 5 min) or subjected to RNA extraction (purified RNA). The viral N1 and N2 genes and the human RNase P gene (RP) were amplified and detected by RT-qPCR. (a) CT values obtained from RT-qPCR analysis of the same samples prepared by the three different methods. (b) Representative amplification curves for each gene obtained for one of the positive samples. (c) Amplification efficiencies (E_PCR) and initial amount of amplicon copies (n) in PK+HID samples relative to the corresponding HID samples. The median of each measurement is represented with a line in the bars and the lengths of these bars represent the standard error. (d) Positive nasopharyngeal swab samples were subjected to treatment with different concentrations of proteinase K (PK) followed by heat inactivation (55°C for 15 min and 98°C for 5 min). The viral N1 and N2 genes and the human RP gene were amplified and detected by RT-qPCR. ΔCT represents the mean difference between CT values obtained in each analyzed condition and the corresponding to HID samples. Mean ± SEM values are represented (N = 5).

Fig 2. PK+HID method exhibits a similar performance than RNA extraction in RT-qPCR determinations of the SARS-CoV-2 N1 gene.

Positive nasopharyngeal swab samples were processed for RNA extraction (purified RNA) or subjected to treatment with PK 1 mg/ml (55°C for 15 min) followed by heat inactivation at 98°C for 5 min (PK+HID). The viral N1 gene and the human RP gene were amplified and detected by RT-qPCR. CT values obtained for the same samples prepared by the two different methods (N = 27), the line obtained by regression of the data (continuous lines) and 95% confidence bands (pink) are represented. The parameter values obtained from the fitting were: slope = 1.17 ± 0.05 and intercept = -3.4 ± 1.2 (N1 amplicon); slope = 1.0 ± 0.1 and intercept = 1.7 ± 3.0 (RP amplicon).

Fig 3. PK+HID method exhibits a similar performance than RNA extraction in RT-qPCR determinations of the SARS-CoV-2 N and ORF1ab genes.

Positive nasopharyngeal swab samples were processed for RNA extraction (purified RNA) or subjected to treatment with PK 1 mg/ml (55°C for 15 min) followed by heat-inactivation at 98°C for 5 min (PK+HID). The viral N and ORF1ab genes and the human RP gene were amplified and detected by RT-qPCR. (a) CT values obtained for the same samples prepared by the two different methods, the line obtained by regression of the data (continuous lines) and 95% confidence bands (pink) are represented. The parameter values obtained from the fitting were: slope = 1.02 ± 0.04 and intercept = 0.1 ± 0.9 (N amplicon); slope = 1.02 ± 0.04 and intercept = 0.8 ± 0.9 (ORF1ab amplicon) and slope = 0.4 ± 0.2 and intercept = 15 ± 6 (RP amplicon). Notice that CT values determined for RP spans a smaller range. (b) Amplification efficiencies obtained in PK+HID samples (E_PK+HID) relative to the corresponding purified RNA samples (E_{purified RNA}). The mean values obtained for E_{purified RNA} were: 1.98 ± 0.03 (N), 2.02 ± 0.06 (ORF1ab) and 1.94 ± 0.03 (RP). The median of each measurement is represented with a line in the bars and the lengths of these bars represent the standard error (N = 16).

Fig 4. Performance of PK+HID method in RT-qPCR determinations of the SARS-CoV-2 N, RdRp and E genes in randomly-selected clinical specimens.

Nasopharyngeal swab samples were processed for RNA extraction (purified RNA) or subjected to treatment with PK 1 mg/ml (55°C for 15 min) followed by heat-inactivation at 98°C for 5 min (PK+HID). The viral N, E and, RdRp genes and the human RP gene were amplified and detected by RT-qPCR. (a) CT values obtained for the same samples prepared by the two different methods, the line obtained by regression of the data (continuous lines) and 95% confidence bands (pink) are represented. RP plot also includes the data obtained in SARS-CoV-2 negative samples and spans a smaller CT range. The parameter values obtained from the fitting were: slope = 1.2 ± 0.1 and intercept = -5 ± 3 (N amplicon); slope = 0.7 ± 0.2 and intercept = 9 ±4 (RdRp amplicon); slope = 0.7 ± 0.1 and intercept = 7 ± 3 (E amplicon) and slope = 0.4 ± 0.1 and intercept = 16 ± 3 (RP amplicon). (b) Amplification efficiencies obtained in PK+HID samples (E_PK+HID) relative to the corresponding purified RNA samples (E_{purified RNA}). The mean values obtained for E_{purified RNA} were: 1.94 ± 0.02 (N), 1.97 ± 0.02 (RdRp), 1.93 ± 0.03 (E) and 1.90 ± 0.03 (RP). The median of each measurement is represented with a line in the bars and the lengths of these bars represent the standard error (N = 8).