key: cord-0685830-5afxlkvs
authors: Capobianchi, M.R.; Rueca, M.; Messina, F.; Giombini, E.; Carletti, F.; Colavita, F.; Castilletti, C.; Lalle, E.; Bordi, L.; Vairo, F.; Nicastri, E.; Ippolito, G.; Gruber, C.E.M.; Bartolini, B.
title: Molecular characterization of SARS-CoV-2 from the first case of COVID-19 in Italy
date: 2020-03-27
journal: Clin Microbiol Infect
DOI: 10.1016/j.cmi.2020.03.025
sha: 75fda70dc20f1d7d650647a03e231110c75a9a06
doc_id: 685830
cord_uid: 5afxlkvs

nan

To the Editor, On January 29, 2020, two Chinese spouses (patient 1, female; patient 2, male), coming to Italy as tourists from Hubei province, were hospitalized at the National Institute for Infectious Diseases "L. Spallanzani", Rome, with fever and respiratory symptoms. SARS-CoV-2 diagnosis was accomplished using real-time RT-PCR [1] on a nasopharyngeal swab and sputum for patient 1 and on a nasopharyngeal swab for patient 2, collected 1 day after symptom onset. Partial sequencing confirmed both patients to be infected with SARS-CoV-2.

A virus isolate was obtained (in a Vero E6 cell line) from the sputum of patient 1, with cytopathic effects evident 24 h postinoculation. At the time of writing, virus isolation from the nasopharyngeal swab sample collected from patient 2 was not successful, likely due to the lower viral load (higher cycle threshold value, 24.56 in the real-time RT-PCR), therefore no further analysis was performed on the virus detected in patient 2. Next-generation sequencing (NGS) was performed on the respiratory samples from patient 1 and on the primary isolate, prior to any further passage, by using the Ion Torrent S5 platform (Thermofisher). The mean count of sequencing reads obtained per sample was 44 000 000 (minimum 41.6 Â 10 6 to maximum 49.7 Â 10 6 ). The reads from the two respiratory samples of patient 1 were merged to obtain a better coverage along the virus genome, and in this paper are referred to as data from the clinical sample. Details of sequencing and bioinformatic analyses are available upon request.

The number of SARS-CoV-2 reads obtained varied from 4079 to >14 Â 10 6 . By using de novo assembly, two contigs of 29 867 nt (mean coverage: 81 324 reads; range: 26e510 718 reads) and 29 792 nt (mean coverage: 80 reads; range: 5e599 reads) were obtained for the isolate and clinical sample of patient 1, respectively, and referred to as consensus sequences. Further analysis was dedicated to identifying the variants present at any nucleotide position for the variability analysis.

Considering the consensus sequences, two non-synonymous changes with respect to the Wuhan-Hu-1 NCBI Reference Genome (Accession number: MN908947.3) [2] were observed in the sequence from the clinical sample from patient 1: G11083T, leading to L3606F change in Orf1a, and G26144T, leading to G251V change in Orf3a. One additional synonymous substitution in Orf1a (A2269T) was detected in the isolate but not in the corresponding clinical sample. All variants were confirmed by Sanger sequencing.

Considering the analysis of genomic variability, several intrasample variants were observed in both the isolate and the clinical sample, but only the positions with a minimum coverage of 20 reads were considered. Intra-sample assessment of overall virus genome variability resulted in 1.27 Â 10 À4 and 1.02 Â 10 À4 nucleotide substitutions per site for the isolate and the clinical sample, respectively. Only two variable positions were observed with a frequency >10% in the clinical sample, both in Orf1a: A2269T (13.73%, coverage: 51x), synonymous for amino acid A668, and G7388A (13.21%, coverage: 53x), leading to amino acid change (A2375T). Interestingly, the frequency of variants at position 2269 was different in the isolate, being T dominant over A in 72% of reads (coverage: 119 582x), accounting for the difference resulting in the consensus sequences.

For the phylogenetic analysis, 87 full-genome SARS-CoV-2 sequences were retrieved from the Global Initiative on Sharing All Influenza Data (GISAID), along with WH-01_MN908947.3 from GenBank. The G26144T substitution observed in the isolate from Italy was also present in five sequences from cases occurring outside of China: EPI_ISL_406596 and EPI_ISL_406597 from France, EPI_ISL_406031 from Taiwan, EPI_ISL_406036 from USA, EPI_-ISL_406844 and EPI_ISL_408977 from Australia. All the genomes carrying this mutation are included in a significant phylogenetic cluster (bootstrap 87%), suggesting a common origin (Fig. 1) ; in fact, the G251V substitution in Orf3a has recently been defined as the marker variant of the 'V' clade (GISAID).

The presence of quasispecies has previously been reported for SARS-CoV and MERS-CoV [3, 4] , suggesting that these betacoronaviruses may consist of complex and dynamic distributions of closely related variants in vivo, similarly to other RNA viruses. When applied to SARS-CoV-2 in this study, the analysis of sequence variability supported the presence of viral quasispecies in the clinical sample as well as in the primary isolate. Namely, two positions with variant frequency >10% were observed in the biological sample, both in Orf1a: A2269T, synonymous, and G7388A, corresponding to amino acid change A2375T. The synonymous variant A2269T, representing a minority variant in the clinical sample, was the dominant one in the isolate. Although low coverage may have affected the precise calculation of minority variant frequency in the clinical sample, the data are consistent with variant selection occurring during the isolation procedure, as previously shown for other respiratory viruses. In the respiratory sample neither mutations nor intra-sample variants were found at positions 8782 and 28 144, recently identified as hotspots of hypervariability (coverage: 61x and 76x respectively) [5] .

Full-genome characterization of new viruses is instrumental for updating diagnostics and assessing viral evolution. On the other hand, virus variability, leading to the development of quasispecies within infected patients, may provide the background for virus evolution and adaptation to new hosts; more studies are necessary to unravel the importance of intra-patient variability in the SARS-CoV-2 evolutionary trajectory.

Genome sequences described on this manuscript are available from GISAID and from GenBank (Acc. Numb: MT008022, MT008023, MT066156 and MT077125).

All authors contributed to the analysis and the writing of the final manuscript.

The authors declare that they have no conflicts of interest. This research was supported by funds to the National Institute for Infectious Diseases 'Lazzaro Spallanzani' IRCCS from the Ministero della Salute, Ricerca Corrente.

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

Genomes collected outside China are highlighted in green. (b) Enlargement of clade reporting SARS-CoV-2/INMI1-Isolate/2020/Italy (in red). Maximum likelihood phylogeny was reconstructed under Hasegawa-Kishino-Yano plus proportion of invariable sites (HKY þ I), inferred by model test function

A new coronavirus associated with human respiratory disease in China

Analysis of intrapatient heterogeneity uncovers the microevolution of Middle East respiratory syndrome coronavirus

SARS-associated coronavirus quasispecies in individual patients

Genomic variance of the 2019-nCoV coronavirus

Letter to the Editor / Clinical Microbiology and Infection xxx (xxxx) xxx

We gratefully acknowledge the contributors of genome sequences of the newly emerging coronavirus, i.e. the originating and submitting laboratories, for sharing their sequences and other metadata through the GISAID Initiative, on which this research is based.