Molecular Characterization of Human Respiratory Syncytial Virus in the Philippines, 2012-2013

Abstract

Human respiratory syncytial virus (HRSV) is a major cause of acute lower respiratory tract infections in infants and children worldwide. We performed molecular analysis of HRSV among infants and children with clinical diagnosis of severe pneumonia in four study sites in the Philippines, including Biliran, Leyte, Palawan, and Metro Manila from June 2012 to July 2013. Nasopharyngeal swabs were collected and screened for HRSV using real-time polymerase chain reaction (PCR). Positive samples were tested by conventional PCR and sequenced for the second hypervariable region (2nd HVR) of the G gene. Among a total of 1,505 samples, 423 samples were positive for HRSV (28.1%), of which 305 (72.1%) and 118 (27.9%) were identified as HRSV-A and HRSV-B, respectively. Two genotypes of HRSV-A, NA1 and ON1, were identified during the study period. The novel ON1 genotype with a 72-nucleotide duplication in 2nd HVR of the G gene increased rapidly and finally became the predominant genotype in 2013 with an evolutionary rate higher than the NA1 genotype. Moreover, in the ON1 genotype, we found positive selection at amino acid position 274 (p<0.05) and massive O- and N-glycosylation in the 2nd HVR of the G gene. Among HRSV-B, BA9 was the predominant genotype circulating in the Philippines. However, two sporadic cases of GB2 genotype were found, which might share a common ancestor with other Asian strains. These findings suggest that HRSV is an important cause of severe acute respiratory infection among children in the Philippines and revealed the emergence and subsequent predominance of the ON1 genotype and the sporadic detection of the GB2 genotype. Both genotypes were detected for the first time in the Philippines.

Fig 1. Temporal distribution of HRSV strains detected in 2012–2013 in the Philippines.

(A) Monthly distribution of HRSV cases in each study site including BPH in Naval City of Biliran Island, EVRMC in Tacloban City of Leyte Island, ONP in Puerto Princesa City of Palawan Island, and RITM in Metro Manila according to genotype. (B) Overall temporal distribution of HRSV strains in four study sites from the period June 2012 to July 2013 in the Philippines.

Fig 2. Phylogenetic analysis of representative HRSV in epidemic season 2012/13.

(A) Phylogenetic tree of HRSV-A based on the nucleotide sequences of the second hypervariable region (2^nd HVR) of the G gene; 342 nucleotides for the 72-nucleotide duplication genotype (ON1 genotype) and 270 nucleotides for the non-duplication genotypes (GA1-7, SAA1, NA1-2 genotypes). (B) Phylogenetic tree of HRSV-B based on the nucleotide sequences of the 2nd HVR of the G gene; 330 nucleotides for the 60-nucleotide duplication genotype (BA genotype) and 270 nucleotides for the non-duplication genotypes (GB1-4, SAB1-4 genotypes). Bootstrap values >70 derived from 1,000 bootstrap replications for evaluating the confidence estimates are shown at branch nodes. The parameters, “all sites” for gap or missing data treatment and “very strong” for branch swap filter, were used in this study. The trees were constructed by adding strains from other countries that were identified by BLASTn search. Genotypes of reference strains from previous studies [4, 36] are indicated in the strain names. Representative strains detected in this study are indicated by the following: NA1 strains (blue circles); ON1 strains (red circles); GB2 strains (purple circles); BA9 strains (green circles). The first ON1 strain that was detected in the Philippines is boldfaced and underlined. Figures inside the parenthesis represent the number of identical sequences.

Fig 3. Deduced amino acids of representative HRSV-A G protein in 2012–2013.

(A) Thirty-one representative unique NA1 amino acid sequences corresponding to position 212 to 301 of the 2^nd HVR of G protein were aligned with the prototype NA1 strain, NG-016-04 (AB470478). (B) Fifty-two representative unique ON1 amino acid sequences corresponding to position 212 to 325 of the 2nd HVR of G protein were aligned with the prototype ON1 strain, ON67-1210A (JN257693). Gray shading for predicted N-linked sites, gray octagons for predicted O-linked sites, dots for identical residues, asterisks for stop codon positions, star for positive selection position, boxes for analogous sites/duplication sites, and parenthesis for number of identical strains are indicated.

Fig 4. Deduced amino acids of representative HRSV-B G protein in 2012–2013.

(A) Two GB2 amino acid sequences corresponding to position 213 to 302 of the 2^nd HVR of G protein were aligned with the prototype GB2 strain, CH93-9b (AF065251). (B) Forty-nine representative unique BA9 amino acid sequences corresponding to position 213 to 322 of the 2nd HVR of G protein were aligned with the prototype BA strain, BA3833/99B (AY333362) and the other prototypes of BA branches. Gray shading for predicted N-linked sites, gray octagons for predicted O-linked sites, dots for identical residues, asterisks for stop codon positions, star for positive selection position, and boxes for analogous sites/duplication sites are indicated.

Fig 5. Shannon entropy plots of deduced amino acid sequences of the 2^nd HVR of G protein of Philippine strains detected in 2012–2013.

Entropy analysis of (A) HRSV-A: NA1 genotype, n = 101 (upper) and ON1 genotype, n = 204 (lower) and (B) HRSV-B: GB2 genotype, n = 2 (upper) and BA9 genotype, n = 116 (lower) in four study sites in the Philippines. The amino acid variability across the second hypervariable region of the G gene is represented by the entropy plot as determined by BioEdit software. The threshold value was set at 0.2. Amino acid sites with entropy values <0.2 are considered conserved and values >0.2 are considered variable.

Fig 6. Phylodynamic analysis of HRSV-A from 2008 to 2013.

Dated phylogeny (upper) and Bayesian skyline plot (lower) of the NA1 and ON1 genotypes. The NA1 genotype was predominant and its population size showed a decreasing trend from 2009 to late 2012 and then exhibited a sharp decline in 2013. The ON1 genotype emerged around late 2012 and became the predominant genotype with an increasing trend in population size in 2013.

Fig 7. Phylodynamic analysis of HRSV-B from 2008 to 2013.

Dated phylogeny (upper) and Bayesian skyline plot (lower) of the BA9 genotype. The population size of the BA9 genotype showed an increasing trend from 2009 and a sharp increase in 2013.

Fig 8. Phylogenetic analysis of GB2 genotype from the Philippines.

The phylogenetic tree was constructed by adding 98 strains from other countries that were identified by BLASTn search. The Philippine GB2 strains may share a common ancestor with strains from Asia.