Detection of a novel coronavirus (SARS-CoV-2) by real-time reverse transcription-polymerase chain reaction, China, 2020

Background: During the outbreak of unexplained pneumonia in the city of Wuhan in the late December, 2019, a novel coronavirus named SARS-CoV-2 was identied as the cause of this outbreak. Methods: A real-time polymerase chain reaction, which targets the orf1ab gene of viral genome, was established to detect and identify the SARS-CoV-2. We used this assay to screen 309 samples from persons with suspected SARS-CoV-2 infection in Wuhan. Then 6 close-phylogenic coronaviruses and 7 viruses which could cause pneumonia were detected. Moreover, 57 clinical samples infected with other viruses and 77 healthy samples were also tested. Results: The limit of detection of the assay was 6.25 copies per reaction in the detection of cRNA transcribed in vitro. The results of detection of throat and fecal swabs from persons with suspected SARS-CoV-2 infection showed throat swabs were more sensitive than fecal swabs during the rst 15 days after onset of symptoms (throat: 56.80%, fecal: 30.43%), while the situation was reversed after 15 days (throat: 20.83%, fecal: 27.58%). And matched pair tests suggested the sputum samples had higher virus loads than throat swabs in the patients (P < 0.05). There was no cross-reaction when we detected the inactive culture of six other coronaviruses (human coronavirus 229E, NL63, OC43, HKU1, SARS-CoV, MERS-CoV) and seven other viruses (inuenza virus A H1N1, inuenza virus A H3N2, inuenza virus B, parainuenza viruses 1, 2, and 3; and respiratory syncytial virus). Besides, 27 BALF samples from pneumonia patients infected with human coronavirus 229E, OC43, HKU1 or human adenovirus 7, 30 throat swabs from patients infected with H1N1 and 77 throat swabs from healthy people tested negative by this assay. Conclusions: The results indicated that the assay specically and sensitively detected the SARS-CoV-2.


Background
Since December 9, 2019, several patients with unexplained pneumonia were found to be epidemiologically associated with Huanan sea food market in Wuhan [1,2], central China, where nonaquatic animals, e.g. birds, snakes and rabbits, were also sold before the outbreak. As of December 31, 2019, a 20 survey identi ed a total of 27 such cases, including 7 severe cases. The number of suspected patients increased to 44 on January 3, 2020 (including 11 severe cases) and to 59 on January 5, 2020 (including 7 severe cases). These patients displayed hyperthermia, some of whom had dyspnea. Chest radiographs showed invasive lesions in double lungs. However, common respiratory viruses, such as human in uenza, avian in uenza, human adenoviruses, severe acute respiratory syndrome coronavirus (SARS-CoV) and middle east respiratory syndrome coronavirus (MERS-CoV), were all detected negative. Metagenomics analysis showed that a novel coronavirus with high homog-enous to bat SARS-like coronavirus maybe the causative agent (unpublished).
Coronaviruses, belonging to the family Coronaviridae, possess a single-stand positive-sense RNA genome with the size ranging between 26-32 kilobases that was the largest genome of all RNA viruses documented thus far. Accumulated evidence shows that many coronaviruses are pathogenic to humans [3][4][5], though most of in-fections just cause mild clinical symptoms. However, two exceptions with severe and even fatal cases caused by coronavirus infections are notable [6,7]. One is SARS-CoV that rst emerged in Guangdong province in southern China in November 2002 and eventually caused > 8,000 human infections and 774 deaths in 37 countries [8,9]. The other case in point is MERS-CoV that was rst detected in the kingdom of Saudi Arabia (KSA) in 2012 [10]. A total of 2494 laboratory-con rmed cases of infection with MERS-CoV have been reported since April 2012 [11], including 858 MERS-CoV associated fatalities.
Until March 9, 2020, the novel coronavirus named SARS-CoV-2 has infected over 109,577 people and caused over 3,809 deaths in 105 countries and territories over the world [12]. Therefore, developing a rapid and accurate diagnostic method for detecting SARS-CoV-2 is an urgent priority for controlling the spread of this infection.
In this study, we describe a sensitive and speci c real-time reverse transcription-polymerase chain reaction (rRT-PCR) assay for the detection of SARS-CoV-2 and report its use in a survey of more than 309 samples from persons diagnosed as probable SARS-CoV-2 infection during the outbreak in Wuhan.

Methods
Design primers and probes Construction of cRNA A calibration standard was generated by diluting RNA transcription of partial orf1a gene of SARS-CoV-2.
The fragment was synthesized based on sequence of orf1a gene of WH04|2020-01-05. The synthesized products were cloned into a pGEM-T Easy vector (Promega Shanghai, Shanghai, China) and then linearized using a speci c DNA restriction enzyme. RNA transcribed in vitro was generated by the linearized plasmid DNA and the RiboMax Express Large-Scale RNA Production System (Promega, Madison, Wisconsin, USA). After digestion of the template DNA with RNase-free DNase I, RNA transcribed in vitro was puri ed with an RNeasy kit (Qiagen GmbH, Hilden, Germany). The puri ed RNA was quanti ed spectrophoto-metrically at 260 nm, divided into aliquots, and stored at -80 °C for future use.
Evaluate cross-reaction

Detection of samples with a diagnosis of suspected SARS-CoV-2 infection
Clinical materials, including 16 sputum, 75 fecal swabs and 218 throat swabs, were obtained by Wuhan BGI medical. All persons had a diagnosis of suspected SARS-CoV-2 infection according to epidemiology data and World Health Organization (WHO) guidance [14]. Total RNA was extracted from 140µL samples using a QIAamp Viral RNA Mini Kit (Qiagen GmbH, Hilden, Germany) in a type II biological safety cabinet.

Results
Sensitivity and speci city of the assay Standard curves of serially diluted RNA transcribed in vitro versus threshold cycle were generated to determine both the e ciency of the rRT-PCR and the limit of detection. The assay exhibited a wide linear range, beginning at 50 copies of target RNA per reaction and extending through 5 × 10 7 copies per reaction (R2 = 0.9995) for the assay (Fig. 1). To determine the detection limit of the assay, two-fold serial dilutions (5 to 0.313 copies/µL) were tested 20 times respectively. The detection rate of approximately 6.25 copies per reaction in the assay were 100% and lower dilutions couldn't be effectively detected.
To evaluate the speci city of our assay, viral RNA of 6 close-phylogenic viruses and 7 viruses which could cause pneumonia were tested, which all showed negative, suggesting no cross-reaction with any of the 13 viruses. Twenty-seven BALF from pneumonia patients infected with other human coronavirus and human adenovirus were tested. Thirty throat swabs from patients infected with H1N1 and 77 throat swabs from healthy people were also tested. All samples showed negative results, while all human samples were positive for human β-actin gene, which was employed as a control.

Results of suspected cases
We analyzed 16 sputum, 75 fecal swabs and 218 throat swabs from probable SARS-CoV-2 infection. The average of detection rate in throat samples was 56.80% (96/169) during the rst 15 days after onset of symptoms ( Fig. 2A). In the 10 days that followed, the detection rate declined to 20.83% (10/49). The average of detection rate in fecal samples was 30.43% (14/46) during the rst 15 days after onset of symptoms ( Fig. 2A). In the 10 days that followed, the detection rate declined to 27.58% (8/29). The results showed the throat swabs is higher sensitive than the fecal swabs in early stage. However, the fecal swabs were higher sensitive than the throat swabs in late stage. The sputum and throat swabs from 16 patients with severe pneumonia were also employed. Fifteen of 16 sputum tested positive with the average of ct value 28.71 ± 4.23. Eleven of 16 swabs tested positive with the average of ct value 34.82 ± 2.57. It seems that the sputum has higher virus loads than throat swabs in the patients (P < 0.05, Fig. 2B).

Discussion
When the dilution of cRNA were tested, the detection limit of the assay in this study showed high sensitivity with approximately 6.25 copies per reaction. When the suspected COVID-19 cases were tested by this assay, different sample types all achieved checkout such as throat swabs, fecal swabs and sputum. The results of these suspected cases suggested that throat swabs were more sensitive than fecal swabs during the rst 15 days after onset of symptoms (throat: 56.80%, fecal: 30.43%), while the situation was reversed after 15 days (throat: 20.83%, fecal: 27.58%). Moreover, matched pair tests of 16 cases indicated that the sputum had signi cantly higher virus loads than throat swabs (P < 0.05).

Conclusions
The results presented here in this study indicated that the assay speci cally and sensitively detected the SARS-CoV-2, which could be considered for the diagnosis of COVID-19 in clinic use. materials and analysis tools. HGW and WJC wrote the paper. All authors have read and approved the nal manuscript. Figure 1 Standard curve and ampli cation plot using serial dilutions of cRNA.