key: cord-0724622-csjowuwt
authors: Sáez-López, Emma; Cristóvão, Paula; Costa, Inês; Pechirra, Pedro; Conde, Patrícia; Guiomar, Raquel; Peres, Maria João; Viseu, Regina; Lopes, Paulo; Soares, Vânia; Vale, Fátima; Fonseca, Patricia; Freitas, Ludivina; Alves, Jose; Pessanha, Maria Ana; Toscano, Cristina; Mota-Vieira, Luísa; Veloso, Rita Cabral; Côrte-Real, Rita; Branquinho, Paula; Pereira‑Vaz, João; Rodrigues, Fernando; Cunha, Mário; Martins, Luís; Mota, Paula; Couto, Ana Rita; Bruges-Armas, Jácome; Almeida, Sofia; Rodrigues, Débora
title: Epidemiology and genetic variability of respiratory syncytial virus in Portugal, 2014–2018
date: 2019-10-10
journal: J Clin Virol
DOI: 10.1016/j.jcv.2019.104200
sha: d4fe065f337e5bfb3a405135d494bf5d12e47e79
doc_id: 724622
cord_uid: csjowuwt

INTRODUCTION: Respiratory syncytial virus (RSV) is associated with substantial morbidity and mortality since it is a predominant viral agent causing respiratory tract infections in infants, young children and the elderly. Considering the availability of the RSV vaccines in the coming years, molecular understanding in RSV is necessary. OBJECTIVE: The objective of the present study was to describe RSV epidemiology and genotype variability in Portugal during the 2014/15-2017/18 period. MATERIAL AND METHODS: Epidemiological data and RSV-positive samples from patients with a respiratory infection were collected through the non-sentinel and sentinel influenza surveillance system (ISS). RSV detection, subtyping in A and B, and sequencing of the second hypervariable region (HVR2) of G gene were performed by molecular methods. Phylogenetic trees were generated using the Neighbor-Joining method and p-distance model on MEGA 7.0. RESULTS: RSV prevalence varied between the sentinel (2.5%, 97/3891) and the non-sentinel ISS (20.7%, 3138/16779), being higher (P < 0.0001) among children aged <5 years. Bronchiolitis (62.9%, 183/291) and influenza-like illness (24.6%, 14/57) were associated (P < 0.0001) with RSV laboratory confirmation among children aged <6 months and adults ≥65 years, respectively. The HVR2 was sequenced for 562 samples. RSV-A (46.4%, 261/562) and RSV-B (53.6%, 301/562) strains clustered mainly to ON1 (89.2%, 233/261) and BA9 (92%, 277/301) genotypes, respectively, although NA1 and BA10 were also present until 2015/2016. CONCLUSION: The sequence and phylogenetic analysis reflected the relatively high diversity of Portuguese RSV strains. BA9 and ON1 genotypes, which have been circulating in Portugal since 2010/2011 and 2011/2012 respectively, predominated during the whole study period.

(92%, 277/301) genotypes, respectively, although NA1 and BA10 were also present until 2015/2016. Conclusion: The sequence and phylogenetic analysis reflected the relatively high diversity of Portuguese RSV strains. BA9 and ON1 genotypes, which have been circulating in Portugal since 2010/2011 and 2011/2012 respectively, predominated during the whole study period.

Respiratory syncytial virus (RSV) A and RSV-B are antigenically different, often co-circulate although one of them usually predominates [1, 2] . Genetic diversity increased with the spread of new genotypes among both of subtypes in last years. The second hypervariable region (HVR2), which carries the C-terminus of the G attachment glycoprotein, has a high degree of divergence and, thereby, it has been used as the main indicator for studies on RSV evolution [3, 4] . To date, based on this region, 15 RSV-A and 29 RSV-B genotypes have been described [3, [5] [6] [7] [8] [9] [10] [11] .

RSV is associated with substantial morbidity and mortality since it is the major cause of lower respiratory tract infections (LRTI) during childhood resulting in hospitalization due to the severity of infection in many cases [12] . Moreover, it causes severe respiratory tract infection in adults, especially elderly, and immunosuppressed patients [13, 14] . Symptomatic supportive care and Palivizumab are currently available options to RSV disease clinical management, which reduce the symptoms and severity, decreasing hospitalization rate but not mortality [15] [16] [17] [18] . Fortunately, RSV vaccines are progressing in phase III clinical trials and could be available in the coming years [19] , being G protein suggested as a plausible target [20] . Therefore, ongoing global surveillance and characterization of circulating RSV strains are required for evidence-based vaccination policies. In Portugal, RSV cases have been detected since 2010 through the national influenza surveillance system (ISS), during the influenza season, from week 37 (September) to week 24 (June) of the next year.

This study aimed to describe the prevalence and genetic variability of RSV during the 2014/15-2017/18 period in Portugal, as well as to evaluate the association between subtype, age and clinical diagnose among patients with laboratory-confirmed RSV infection.

The Portuguese ISS accounts with the sentinel and non-sentinel components. Sentinel ISS is made up of General Practitioners' (GP) Sentinel Network and GP from the EuroEva Portuguese component of the I-MOVE project [21] , and the Emergency/Obstetric Departments Networks in hospitals. Non-sentinel ISS integrates the Portuguese Laboratory Network for the Diagnosis of Influenza Infection (PLNDII), which comprises 14 hospital-based laboratories and is coordinated by the National Institute of Health Doutor Ricardo Jorge (INSA). This network is integrated by general hospitals with pediatric and adults emergency rooms and medical wards, and one reference pediatric hospital in Lisbon. ISS weekly reports laboratory data of tested samples to the European Influenza Surveillance Network, which is coordinated by the European Centre for Disease Prevention and Control (ECDC) and WHO/Europe.

Demographic, clinical and laboratory data from RSV-positive cases were collected through the ISS. RSV-positive respiratory samples were collected from three populations: (i) all age influenza-like illness (ILI) [22] patients reported by the sentinel ISS during the 2014/15-2017/18 period; (ii) children aged < 5 years with respiratory infection diagnosed by the PLNDII during the 2015/16-2017/18 period; and (iii) adults aged≥65 years with respiratory infection diagnosed by the PLNDII during the 2017/18 season. Samples included nasopharyngeal or oropharyngeal swabs, and nasopharyngeal aspirates and lavages. This study was approved by the Health Ethic Committee of INSA.

Viral nucleic acid extraction was performed using the automated acid extraction platform EasyMAG. Molecular detection of RSV and identification of the subtypes A and B were performed using real-time PCR adapted from a previous protocol described by Gunson et al. [23] .

For genotyping, sequencing of RSV-positive ISS cases was prospectively performed since the 2015/16 season and retrospectively carried out before that time. In addition, the PLNDII voluntarily selected the first two RSV positive samples per week from children aged < 5 years (2015/16-2017/18) and adults aged ≥65 years (2017/ 18). Conventional PCR and sequencing analysis of the HVR2 was performed by a one-step RT-PCR adapted from previously published protocols. For the first-round PCR, RSV3 (GGCAATGATAATCTCAAC) [24] and F164 (GTT ATG ACA CTG GTA TAC CAA CC) [25] were used, whereas for the second round PCR and sequencing, OGCH496+ (GAT GATTACCATTTTGAAGTGTTCA) [25] and F1 (CAACTCCATTGTTATT TGCC) [26] were selected. The size of the product was around 500 pb.

Potentially N-glycosylation sites (Asn-X-Ser/Thr) were identified by NetNGlyc 1.0. server (http://www.cbs.dtu.dk/services/NetNGlyc/). Alignments were performed with the MUSCLE method and phylogenetic trees were generated using the Neighbor-Joining method and pdistance model (bootstrap re-sampling of 1000 replicates) on MEGA 7.0. In addition to worldwide reference strains (RSV-A and RSV-B) that were retrieved from GenBank (https://www.ncbi.nlm.nih.gov/ genbank/), 20 RSV-A strains from a previous study in Portugal (2010/11-2013/14) [27] were used for the tree construction together with the sequences generated in this study. Sequences with 100% homology in the nucleotide sequence were represented by one strain and submitted to GenBank database with accession numbers MN122441-MN122562 (RSV-A) and MN122563-MN122694 (RSV-B).

Demographic and epidemiological characteristics were described using proportions, and group comparisons were assessed through Chisquare test or Fisher's exact test. Tests with p-value < 0.05 were considered statistically significant. The statistical analysis were performed using STATA 12.

RSV was identified in 3235 samples through the ISS from week 37 of 2014 to week 24 of 2018 ( Fig. 1) . A total of 407 (12.6%) RSV cases were co-detected with other respiratory viruses, mainly Picornavirus (29.5%, 120/407), Coronavirus (11.5%, 47/407), and Adenovirus (10.1%, 41/407). The majority of RSV cases (97%, 3138/3235) were y, years. * Cases with missing age or sex included in the study. Age was missing for five RSV-positive and 65 RSV-negative cases; sex was missing for one RSVpositive and 13 RSV-negative cases. ** Observed frequencies where higher than those expected among the 0-4 years age group (P < 0.0001). 

Demographic characteristics for RSV-positive and negative cases were described (Table 2) . RSV-positive frequency (72.5%, 2343/3230) was significantly (P < 0.0001) higher among children aged < 5 years than the one expected if there was not association between age and RSV infection. The sequence of the HVR2 was determined for 542 RSV-positive samples during the 2014/15-2017/18 period. Demographic and clinical characteristics for those 542 patients, including 56 through the sentinel ISS (Table 3 ) and 486 through the PLNDII (Table 4) , were described and the association with the RSV subtype was assessed. On RSV cases detected through the ISS, observed RSV-B subtype frequency (86.4%, 19/22) was significantly higher in females (P = 0.012) and in 15-44 years age group (P = 0.037). Samples from RSV cases that were detected through the PLNDII, belonged mostly to children (88.3%, 429/ 486) and 59.9% (291/486) were aged < 6 months. They presented a RSV-A frequency (66.2%, 137/207) significantly (P = 0.015) higher than the one expected. Three hundred thirty-six (69.1%) cases were hospitalized patients. Bronchiolitis in children aged < 6 months (62.9%, 183/291) and ILI in adults aged ≥65 years (24.6%, 14/57) were found to be significantly (P < 0.0001) associated with RSV laboratory confirmation. Moreover, observed frequencies of RSV-positive ILI (8.1%, 6/74) and pneumonia (10.8%, 8/74) among children from 13 months to 4 years old were also significantly (P = 0.002) higher than the ones expected. Therefore, these diagnostics were associated with laboratory-confirmed RSV infection among those age groups.

In addition to the 542 samples from the studied period, 20 RSV-A strains from the 2010/14 period were included in the phylogenetic analysis ( (Fig. 2) . All RSV-B strains clustered in two Buenos Aires genotypes, BA9 (92%, 277/301) and BA10 (8%, 24/301), the latter being only present until the 2015/16 season (Fig. 3) .

The HVR2 of all NA1 strains was aligned with the reference NA1 strain (AB470478) whereas ON1 strains were compared with the reference ON1 strain (JN257693) (Supplementary Fig. 1 ). Eight NA1 and 114 ON1 strains showed at least one different nucleotide, being the highest number of amino acid substitutions found in the ON1 genotype. I243S, E262 K, L274 P, L298 P, and Y304H were the most common substitutions among ON1 strains. In addition, they contained E232 G/ R/K and T253 K/R substitutions in comparison to NA1 strains and, in fact, the last substitution led to loss of one potential N-glycosylation site. Another site loss was due to N273Y/H/S substitution in all strains with the exception of one NA1 strain (2010/11). However, in addition to two conserved N-glycosylation sites (N237 and N318), two novel sites due to I243 N and S299 N substitutions were observed in one and two ON1 strains, respectively.

The HVR2 of all RSV B strains was analyzed in comparison to the first reported BA strain (AY333364), which belongs to BA1 genotype ( Supplementary Fig. 2) . A total of 132 RSV-B group strains, including six BA10 and 126 BA9 strains with 98 different amino acidic combinations, were observed. Substitutions K218 T, L223 P, S247 P, T270I, V271A, I281 T and H287Y were identified in most RSV-B viruses. All BA10 strains showed the same amino acid sequence and had the substitution E292 G, which was exclusive of this genotype. Three conserved sites were identified at amino acid 230, 296 and 310, although few BA9 strains lost N296 and N310 due to different substitutions. However, two additional sites (amino acid substitutions T227 N and H279 N) were identified in specific RSV-B strains.

Circulating Portuguese RSV strains, through sentinel and non-sentinel ISS between 2014 and 2018, were characterized in epidemiologic and genetic terms in order to provide a better understanding of the molecular epidemiology of RSV infection in the country.

RSV detection rate in the sentinel ISS (2.5%) was similar to frequencies reported in outpatients from Brazil [28] , Korea [29] and China [30] although was lower than in other European countries [15, 31] . In the PLNDII, RSV prevalence was higher than the one found in the sentinel ISS since this network is composed by hospital-based laboratories, collect data from inpatients and sentinel ISS use the European Union (EU) ILI case definition [22] , which is not accurate for RSV detection in Portugal [32] . The remarkable number of tested samples (8212) during the 2017/18 season could be explained with the consciousness-raising of relevance of RSV disease among physicians. Therefore, from our point of view, the prevalence of this season (14.4%) might be the best representative in the country, being similar to findings from Senegal [33] , Thailand [34] , Spain [35] and different regions in India [36, 37] . Nonetheless, a true comparison of RSV prevalence in different geographic areas is difficult to perform taking into account the differences in the study design and, even in our study, prevalence was wide-ranging according to the season. Bronchiolitis and pneumonia were significantly associated with RSV-positive cases among children aged < 5 years, which was consistent with previous studies that found RSV as a major cause of acute LRTI in this age group [12, [38] [39] [40] [41] .

RSV-A and RSV-B subtypes co-circulated between 2010 and 2018 in Portugal [27, 42] . RSV-A subtype was the most prevalent in the 2015/ 16 season, whereas RSV-B subtype dominated in the next two consecutive seasons (2016/17 and 2017/2018). Periodic shifts (1-3 years) in RSV subtype have been reported all over the world [40] . RSV-A subtype was significantly higher in children younger than 5 years and especially those aged below 6 months, which may be concerning since RSV-A infection has been associated with more severe bronchiolitis and higher disease severity scores [43, 44] . However, discrepancies have been described [45] , and indeed, specific substitutions may indicate the most virulent strains during a specific period of time better than the subtype [46] . RSV-B subtype was significantly higher in 15-44 years old and female patients. However, this fact could be explained due to the low number characterized cases through the sentinel ISS. RSV-A strains (between 2010 and 2018) clustered in NA1 and ON1 genotypes, which was consistent with several studies reporting the almost exclusivity of these genotypes after 2005 [4, 47] . NA1 genotype was detected from 2010 to 2012, and during the 2015/16 season. However, the fact that this genotype was not found between these two periods might be due to the sequencing bias. In Portugal, ON1 genotype was detected during the 2011/12 season for the first time [27] , and since then, it has predominated in the country as reported worldwide [48, 49] . Regarding RSV-B genetic diversity, all belonged to BA genotype, which has already been detected in at least 23 countries [50] . BA9 genotype has predominated in Portugal since 2010 [27, 42] as in other countries [10, 45, 49, 51] , being detected alongside with BA10 genotype until 2015/16, when BA9 remained the only genotype detected onwards. Since 2001, new genotypes have been added over time by different authors following the proposed criteria by Venter et al. [5] . However, the addition of new sequences from different locations or more recent epidemics led to re-classify previous sequences into an existing genotype or even consolidate a new one [9, 26, 52, 53] . In the present study, we considered that all RSV-A non-NA1 strains belonged to ON1 genotype since mostly of the identified clusters showed a bootstrap value < 70% and had the 72-nc insertion characteristic from this variant. Therefore, a standardized criteria may play a key role to assign an adequate genotype and assure the comparability between different studies regardless the addition of new sequences.

The alignment of the predicted amino acid sequences of strains from BA9, NA1 and ON1 genotypes confirmed the relatively high genetic variability associated with the HVR2. Strains from both of subtypes showed common changes, conserved N-glycosilation sites and subtitutions leading to loss or gain of these sites that had been previously reported [11, 40, 41, 45, 46, 49, 50, [54] [55] [56] [57] [58] [59] . Within RSV-A strains, the number of substitutions among ON1 strains was higher compared with its NA1 counterparts. BA10 strains showed 100% homology in the HVR2 to reference strains from Thailand (KY328142,KY328145) and had the exclusive substitution E292 G, which has been observed in other studies [58, 59] . Continual surveillance of amino acid changes as well as novel N-linked glycosylation sites should be performed since alterations in the antigenic characteristics of the viral surface glycoproteins may enable immune evasion leading to an evolutionary advantage [45, 58] . In fact, the spread of BA genotype has been explained by the 60-nc duplication in the HVR2, which was associated with an increase of virus attachment and fitness [60] .

The main limitation of this study was being restrospectively performed before the 2015/16 season. Consequently, there was a low number of successfully sequenced RSV samples before that time and through the sentinel ISS due to insufficient viral load or RNA degradation during conservation. However, the study had the advantage of including a large sample size from different regions during four consecutive seasons in Portugal, and thereby, it provides useful information on RSV epidemiology and genotype variability in the country from a recent period.

In conclusion, this study was the first to explore RSV epidemiology and genotype variability in Portugal from 2014 to 2018. It constitutes a contribution to RSV knowledge on evolution and strain circulation that is necessary for RSV vaccine development and future monitoring of vaccine effectiveness. Continuous surveillance of specific genomic changes and a standardized criteria for an optimized genotype assignment would be relevant as regards the vaccine design and RSV surveillance including virus characterization.

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Ethical approval: This study was approved by Health Ethic Committee of INSA. 

None.

Two distinct subtypes of human respiratory syncytial virus

Respiratory syncytial virus-a comprehensive review

A study of the genetic variability of human respiratory syncytial virus (HRSV) in Cambodia reveals the existence of a new HRSV group B genotype

Conservation of G-protein epitopes in respiratory syncytial virus (Group A) despite broad genetic diversity: is antibody selection involved in virus evolution?

Genetic diversity and molecular epidemiology of respiratory syncytial virus over four consecutive seasons in South Africa : identi cation of new subgroup A and B genotypes

Genetic variability of respiratory syncytial viruses (RSV) prevalent in Southwestern China from 2006 to 2009: emergence of subgroup B and A RSV as dominant strains

Prevalence and genetic characterization of respiratory syncytial virus (RSV) in hospitalized children in Korea

Genetic variation in attachment glycoprotein genes of human respiratory syncytial virus subgroups A and B in children in recent five consecutive years

Molecular characterization of human respiratory syncytial virus, 2010-2011: identification of genotype ON1 and a new subgroup B genotype in Thailand

Circulation of a novel human respiratory syncytial virus Group B genotype during the 2014-2015 season in Catalonia (Spain)

Genetic variability of human respiratory syncytial virus group B in Panama reveals a novel genotype BA14

Global, regional, and national disease burden estimates of acute lower respiratory infections due to respiratory syncytial virus in young children in 2015: a systematic review and modelling study

Seasonality and geographical spread of respiratory syncytial virus epidemics in 15 European countries

Global disease burden estimates of respiratory syncytial virus-associated acute respiratory infection in older adults in 2015: a systematic review and meta-analysis

Workshop on burden of RSV disease in Europe, ECDC Expert Consultation Meeting

Atualização de norma DGS: prescrição de Palivizumab para prevenção de infeção pelo vírus sincicial respiratório em crianças de risco | a enfermagem e as leis

Updated guidance for palivizumab prophylaxis among infants and young children at increased risk of hospitalization for respiratory syncytial virus infection

World Health Organization, Preferred product characteristics for respiratory syncytial virus (RSV)

Ongoing developments in RSV prophylaxis: a clinician's analysis

Respiratory syncytial virus: targeting the G protein provides a new approach for an old problem

on the Communicable Diseases and Related Special Health Issues to Be Covered by Epidemiological Surveillance As Well As Relevant Case Definitions

Real-time RT-PCR detection of 12 respiratory viral infections in four triplex reactions

Molecular epidemiology and genetic variability of respiratory syncytial virus (RSV) in Stockholm

Ten years of global evolution of the human respiratory syncytial virus BA genotype with a 60-Nucleotide duplication in the G protein gene

Circulation patterns of genetically distinct group A and B strains of human respiratory syncytial virus in a community

Molecular epidemiology of respiratory syncytial virus between 2010-2015, in Portugal

Acute respiratory infection and influenza-like illness viral etiologies in Brazilian adults

Laboratory surveillance of influenza-like illness in seven teaching hospitals

Human respiratory syncytial virus in children with lower respiratory tract infections or influenza-like illness and its co-infection characteristics with viruses and atypical bacteria in Hangzhou

Use of the moving epidemic method (MEM) to assess national surveillance data for respiratory syncytial virus (RSV) in the Netherlands

Evaluation of ECDC Influenza-like illness (ILI) case definition to detect respiratory syncytial virus (RSV) infection through the Influenza Surveillance System in Portugal

Epidemiology and molecular characterization of human respiratory syncytial virus in Senegal after four consecutive years of surveillance

Respiratory syncytial virus, human metapneumovirus, and influenza virus infection in Bangkok

Virological surveillance of human respiratory syncytial virus A and B at a tertiary hospital in Catalonia (Spain) during five consecutive seasons

Prevalence and molecular characterization of circulating respiratory syncytial virus (RSV) in Chennai, South India during

Genetic characterization respiratory syncytial virus in Kerala, the southern part of India

Novel respiratory syncytial virus (RSV) genotype ON1 predominates in germany during winter season 2012-13

Respiratory syncytial virus subtype ON1/NA1/BA9 predominates in hospitalized children with lower respiratory tract infections

Prevalence and genetic characterisation of respiratory syncytial viruses circulating in Bulgaria during the 2014/15 and 2015/16 winter seasons

The significance of human respiratory syncytial virus (HRSV) in children from Ghana with acute lower respiratory tract infection: a molecular epidemiological analysis

Molecular characterization of respiratory syncytial virus during 2015-2016 season in Portugal

Does respiratory syncytial virus subtype influences the severity of acute bronchiolitis in hospitalized infants?

Human respiratory syncytial virus subtypes A and B infection among children attending primary and secondary health care facilities in Ibadan

Characteristics and their clinical relevance of respiratory syncytial virus types and genotypes circulating in Northern Italy in five consecutive winter seasons

High epidemic burden of RSV disease coinciding with genetic alterations causing amino acid substitutions in the RSV G-protein during the 2016/ 2017 season in the Netherlands

Circulation of HRSV in Belgium: from multiple genotype circulation to prolonged circulation of predominant genotypes

Rapid replacement of human respiratory syncytial virus A with the ON1 genotype having 72 nucleotide duplication in G gene

Genetic characterization of respiratory syncytial virus highlights a new BA genotype and emergence of the ON1 genotype in Lyon

BA9 lineage of respiratory syncytial virus from across the globe and its evolutionary dynamics

A molecular epidemiological study of human respiratory syncytial virus in Croatia

Antigenic and genetic variability of human respiratory syncytial viruses (group A) isolated in Uruguay and Argentina

Genetic variability among complete human respiratory syncytial virus subgroup A genomes: bridging molecular evolutionary dynamics and epidemiology

Genetic diversity of human respiratory syncytial virus circulating among children in Ibadan

Detection of ON1 and novel genotypes of human respiratory syncytial virus and emergence of palivizumab resistance in Lebanon

Molecular characterization of human respiratory syncytial virus in the Philippines

Genetic variability of human respiratory syncytial virus a strains circulating in Ontario: a novel genotype with a 72 nucleotide G gene duplication

Respiratory syncytial virus genotypes NA1, ON1, and BA9 are prevalent in Thailand

New genotypes within respiratory syncytial virus group B genotype BA in Niigata

Functional analysis of the 60-nucleotide duplication in the respiratory syncytial virus Buenos Aires strain attachment glycoprotein

We acknowledge the contribution of all the physicians that collaborated in the Portuguese Influenza Surveillance Program that are members of the Sentinel GP Network, of the Emergency and Obstetric units network and from the EuroEva project. Moreover, we would like to thank to the PLNDII. In addition, this material is based upon work supported by the Fellowship European Public Health Microbiology Training (EUPHEM), ECDC.

Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.jcv.2019.104200.