key: cord-0702564-jpli9msr
authors: Nonaka, Carolina K. V.; Franco, Marília Miranda; Gräf, Tiago; de Lorenzo Barcia, Camila Araújo; de Ávila Mendonça, Renata Naves; de Sousa, Karoline Almeida Felix; Neiva, Leila M. C.; Fosenca, Vagner; Mendes, Ana V. A.; de Aguiar, Renato Santana; Giovanetti, Marta; de Freitas Souza, Bruno Solano
title: Genomic Evidence of SARS-CoV-2 Reinfection Involving E484K Spike Mutation, Brazil
date: 2021-05-03
journal: Emerg Infect Dis
DOI: 10.3201/eid2705.210191
sha: 301640630c55529765df37376435af69e0122312
doc_id: 702564
cord_uid: jpli9msr

Uncertainty remains about how long the protective immune responses against severe acute respiratory syndrome coronavirus 2 persists, and suspected reinfection in recovered patients has been reported. We describe a case of reinfection from distinct virus lineages in Brazil harboring the E484K mutation, a variant associated with escape from neutralizing antibodies.

V iral evolution might favor reinfections (1), and the recently described spike mutations, particularly in the receptor binding domain in severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) lineages circulating in the United Kingdom, South Africa, and most recently in Brazil (A. Rambaut et al., unpub. data, https://virological.org/t/preliminary-genomic-characterisation-of-an-emergentsars-cov-2-lineage-in-the-uk-defined-by-a-novelset-of-spike-mutations/563; H. Tegally et al., unpub. data, https://doi.org/10.1101/2020.12.21.20248640 ; C.M. Voloch et al., unpub. data, https://doi.org/ 10.1101/2020.12.23.20248598), have raised concern on their potential impact in infectivity, immune escape, and reinfection. We report a case of reinfection from distinct SARS-CoV-2 lineages in Brazil harboring the E484K mutation, a variant associated with escape from neutralizing antibodies (2; A.J. Greaney, unpub. data, https://doi. org/10.1101/2020.12.31.425021; Z. Liu, unpub. data, https://doi.org/10.1101/2020.11.06.372037).

A 45-year-old woman residing in Salvador (Bahia State, northeast Brazil) with no underlying conditions had symptoms of viral infection on 2 occasions (May 26, 2020, and October 26, 2020). In the first episode, the patient had diarrhea, myalgia, asthenia, and odynophagia for ≈7 days. She took 40 mg prednisone for 5 days and returned to normal activities 21 days later, after resolution of symptoms without sequelae or complaints. In the second episode, which was symptomatically more severe in terms of intensity and duration, the patient had headache, malaise, diarrhea, cough, and sore throat that evolved to myalgia and ageusia, muscle fatigue, insomnia, mild dyspnea on exertion, and shortness of breath. In both episodes, however, disease was classified as mild, and she was treated at home, not requiring hospitalization.

The patient was a healthcare executive. Identified workplace exposure included frequent meetings with coronavirus disease (COVID-19) frontline physicians and healthcare teams. Also, before the second episode, she attended a meeting with a group of physicians, one of whom had COVID-19 diagnosed in the days following.

On both occasions, viral RNA was extracted from nasopharyngeal swab specimens and tested for SARS-CoV-2 by multiplex real-time reverse transcription PCR (rRT-PCR) Allplex SARS-CoV-2 assay (Seegene, https://www.seegene.com). Both times, results of rRT-PCR tests targeting 3 genes (N, E, and RdRp) were positive for SARS-CoV-2 ( Figure, panel  A) . Cycle threshold values of N, E, and RdRp targets were 25, 26, and 27 in the first episode and 21, 12, and 17 in the second episode, respectively. In the second episode, the patient had a high viral load (presumed because of low cycle threshold values detected). Four weeks after the patient tested positive by rRT-PCR in the second episode, an IgG test against S1 protein by chemiluminescence (VITROS, Ortho Clinical Diagnostics, https://www.orthoclinicaldiagnostics.com) yielded a positive result. We then sequenced swab specimens by using PGM Ion Torrent (ThermoFisher, https://www.thermofisher.com), according to the manufacturer's instructions. A total of 1,405,009 mapped reads for sample A (from the first episode) and 2,570,182 reads for sample B (from the second episode) were obtained, resulting in a sequencing mean depth >1,000× for both samples and a coverage of >99%.

We further assessed the distinct viral origin of the 2 infections by phylogenetic inference, comparing the 2 new isolates (GISAID accession nos. EPI_ ISL_756293 and EPI_ISL_756294; https://www.gisaid.org) with all SARS-CoV-2 genomes from Brazil available through GISAID as of January 14, 2021. Only genomes >29,000 bp and <1% of ambiguities were retrieved (n = 1,164). Sequences were aligned by using MAFFT (3) and submitted to IQ-TREE for maximum-likelihood phylogenetic analysis (4). We inferred time-scaled trees by using TreeTime (5) .

Comparison of the phylogenetic profiles of the 2 new sequences with contemporaneous sequences from Brazil (Appendix Table, https://wwwnc.cdc. gov/EID/article/27/5/21-0191-App1.xlsx) clearly demonstrated that the 2 COVID-19 episodes, separated by a 147-day interval, were indeed caused by different SARS-CoV-2 lineages, confirming reinfection (Figure, panel B) . In the first episode, the lineage B.1.1.33 was detected, whereas lineage P.2 (an alias for B.1.1.28.2) was detected in the second infection ( Figure, panel B) https://virological.org/t/genomic-characterisation-of-an-emergent-sars-cov-2-lineage-in-manauspreliminary-findings/586); and P.2, also described in Brazil (C.M. Voloch et al.) and already detected in the United Kingdom, United States, Canada, and Argentina (https://cov-lineages.org/lineages.html). Our report of SARS-CoV-2 reinfection with a E484K variant corroborates in vitro and in silico studies that estimated the potential of lineages carrying this mutation to escape from neutralizing antibodies (3; Z. Liu et al.) and highlights the importance of genomic surveillance to detect and monitor the emergence of new viral linages with possible implications for public health policies and immunization strategies.

Genomic evidence for reinfection with SARS-CoV-2: a case study

Escape from neutralizing antibodies by SARS-CoV-2 spike protein variants

Parallelization of MAFFT for large-scale multiple sequence alignments

IQ-TREE: a fast and effective stochastic algorithm for estimating maximum-likelihood phylogenies

TreeTime: Maximumlikelihood phylodynamic analysis

We thank the personnel from Health Surveillance System from the São Rafael Hospital that helped with samples, sources, epidemiological data collection. We also would like to thank all the authors who have kindly deposited and shared genome data on GISAID. A table with genome sequence acknowledgments is available at https://github. com/genomicsurveillance/SARS-CoV-2-surveillance. 

Dr. Nonaka is a researcher at the Biotechnology and Cell Therapy Centre of São Rafael Hospital, Salvador, Brazil. Her research focuses on the implementation of next-generation sequencing techniques to track the spread of emerging and reemerging viral pathogens and to identify novel disease biomarkers.