Methods

Sample collection

This study was carried out as part of the national influenza surveillance programme in CAF. Samples were collected from children under 5 years of age who came to the Bangui Paediatric Complex for consultation for a severe acute respiratory illness (SARI) syndrome or to four other sentinel sites for an influenza-like illness (ILI) syndrome. From January 2015 to December 2018, a total of 3903 nasopharyngeal swabs were collected from patients meeting the WHO criteria for ILI (measured fever or history of fever ≥38°C and cough with onset within the last 10 days) or SARI (measured fever or history of fever ≥38°C and cough with onset within the last 10 days requiring hospitalization) (Fitzner et al., 2018). The swab was placed in a labelled tube with 3 mL of Universal Transport Medium (UTM; Copan, Italy) and stored at 4°C or sent immediately in a cold box to the National Reference Centre (NRC) for Influenza for diagnosis. Each sample was then aliquoted into four 1.5 mL Eppendorf tubes for influenza virus culture, extraction of nucleic acids and long-term storage at -80°C. Social, demographic, clinical and epidemiological data were recorded for each patient using a standardised questionnaire.

Nucleic acid extraction and RSV detection by RT-PCR

RNA was extracted from a 140 µL nasopharyngeal swab sample aliquot using the QIAamp Viral RNA Mini kit (QIAGEN, USA) according to the manufacturer’s protocol. RSV detection was performed by a conventional multiplex one-step RT-PCR detecting human metapneumovirus, influenza A and influenza B virus as described before 21. Cycling conditions were as follows: 50°C for 30 min, 94°C for 15 min followed by 40 cycles of 94°C for 30 sec, 55°C for 30 sec, 72°C for 1 min, and a final extension at 72°C for 10 min.

RSV typing and sequencing

RSV positive samples were then typed as RSV-A or -B by targeting fusion protein F by a semi-nested PCR as described before 22. cDNA synthesis was carried out using the SuperScript III First-Strand enzyme (Invitrogen, USA) and the anti-sense primer F164 (5’-GTTATGACACTGGTATACCAACC-3’) 22.
The first round amplification was carried out with 5 U/µL Taq DNA polymerase (Promega, USA) and 10 µM of each primer, ABG490 (5’-ATGATTWYCYTTTGAAGTGTTC-3’) and F164. Two semi-nested PCRs were then performed on each sample with the AG655 (5’-GATCYCAAACCTCAAACCAC-3’) or BG517 (5’-TTYGTTCCCTGTAGTATATGTG-3’) sense primers and F164 antisense primer 22.
PCR products were separated by electrophoresis on a 1.5% agarose gel. Samples with a 450 bp band in the first semi-nested PCR were typed as RSV-A and samples with a band between 585 pb and 645 bp in the second semi-nested PCR were typed as RSV-B.  These amplicons were then purified using the QIAquick PCR Purification kit (QIAGEN, Venlo, the Netherlands) according to the manufacturer’s protocol. Purified products were sequenced using the corresponding PCR primers on an ABI 3130 capillary sequencer.

Data availability

Unique study sequences were deposited in GenBank under the accession numbers xxx to xxx.

Phylogenetic analysis

Consensus sequences were generated using SeqScape software (version 2.5) and analysed with BLAST (http://www.ncbi.nlm.nih.gov.BLAST/) for similarity searches. All sequences were aligned using ClustalW implemented in BioEdit v7.2.5 23. Evolutionary distances were calculated using the Maximum Composite Likelihood model and are expressed as the number of nucleotide substitutions per site. The best nucleotide substitution model was selected with MEGA v6.0624, and used to calculate the phylogenetic trees with the Maximum Likelihood method with 1000 bootstrap iterations as implemented in MEGA v6.06.

Amino-acid analysis

The nucleotide sequences were translated into amino-acid sequences using the standard genetic code implemented in BioEdit software. The amino-acid sequences were compared with those from the prototype RSV-A (ON1: JN257693, Canada; NA1: AB470478, Canada) and RSV-B strains (BA9: AY333364, Argentina). BioEdit was used to visualise the amino-acid variability in the second hypervariable part of the G protein.

Glycosylation sites

Potential glycosylation sites in the second hypervariable part of the G protein were identified using NetNGlyc 1.0 server (http://www.cbs.dtu.dk/services/NetNGlyc) and NetOGlyc 4.0 server (http://www.cbs.dtu.dk/services/NetOGlyc). Oxygen (O)-linked glycosylation sites occur on serine (S) and threonine (T) and nitrogen (N)-linked glycosylation sites occur on N-X-S/T, where X is any amino acid other than proline (P).

Statistical analyses

Statistical analyses were performed in SigmaPlot v12.5, using Mann-Whitney rank sum tests for continuous variables, χ2-test for categorical variables or z-test for low proportions. Odds ratios were calculated with χ2-test with the Yates correction.

Ethical considerations

The surveillance programme carried out in CAF was approved by an ethics committee composed of experts from the Ministry of Health (Decree 0277/MSPP/CAB/DGSPP/DMPM/SMEE of 5 August 2002). Participants were only included in the study after obtaining verbal consent from the child’s parents or legal guardians. Data were pseudo-anonymised, in strict compliance with patient privacy rights. The results were sent to the child’s regular medical practitioner in a sealed envelope.