Detection of respiratory viruses
We have described detection rates of several respiratory viruses from September 2017 to December 2021 in this study. A total of 6499 specimens were tested through Immunofluorescence assay for preliminary screening of respiratory viruses. Overall detection rate for all respiratory viruses from total collected specimens was 16.12% (1048 /6499). A total of 405 specimens (6.20%, 405/6499) were found positive for RSV. Among respiratory viruses, the highest detection rate was of RSV (6.20%, 405/6499) followed by IFA 2.8% (180/ 6499), AdV 2.13% (139/6499), PIV3 1.9% (123/1048), IFB 1.6% (101/6499) PIV1 1.3% (83/6499) and PIV2 (0.3%) (19/6499) as shown in Table 1. Among total positive specimens (n = 1048), detection rate for RSV, IFA, AdV, PIV3, IFB and PIV2 was 38.45%,17.17%, 13.26%, 11.74%, 9.64%, 7.92% and 1.81% respectively (Table 2). Out of 405 RSV positive specimens, 93 specimens were selected for RSV subtyping, genotyping, and phylogenetic analysis. Noteworthy, all RSV positive samples were found negative for coinfection with IFA, IFB, AdV, PIV1, PIV2 and PIV3. Our preliminary analysis revealed higher incidence of RSV in children especially under 6 months of age, mostly showing clinical signs of fever and cough. The median duration of hospital stay was seven days for children suffering from RSV infections in this study. Unfortunately, we could not collect complete information about clinical parameters, therefore, these findings are not presented in detail in this study.
Seasonal distribution of respiratory viruses: We collected and analyzed data about seasonal distribution of respiratory viruses. To analyze seasonal distribution of RSV, we divided Guangdong province weather into four seasons: spring (March, April, and May), summer (June, July and August), autumn (September, October and November) and winter (December, January, and February) as described previously [34]. There were statistical associations between respiratory viruses (IFA, IFB, AdV, RSV, PIV1, PIV2 and PIV3) and months (p < 0.05 or p > 0.05). Yearly distribution of RSV cases has been presented in Fig. 2. The annual RSV incidences were 5.8% (119/6499), 5.3% (103/6499),11.17% (80/6499), and 6% (72/6499) from 2018 to 2021, respectively (Table 2). We did not include seasonal data for 2017 as sampling data was not enough for seasonal distribution comparison. Notably, we found significantly higher RSV incidences in the year 2020 (11.17%, p < 0.05).
As shown in Fig. 2, overall detection rate for respiratory viruses in 2018 was 19.2%, 11.2%,17.5% and15.5% in winter, spring, summer, and autumn, respectively. While RSV detection rate in 2018 was 2.7%, 5.6%, 9.4% and 5.3% in winter, spring, summer, and Autumn, respectively (Fig. 3). Surprisingly, RSV detection rate in summer was the highest in 2018 (9.4%, p < 0.05). Overall detection rate of respiratory viruses in 2019 was 29.5%, 21.7%, 17.7% and 11% in winter, spring, summer, and autumn, respectively. While RSV detection rate in 2019 was 8.6%, 7.5%, 2.5% and 2% in winter, spring, summer, and autumn, respectively. As expected, RSV detection rate in winter was the highest in 2019 (8.6%, p < 0.05). In 2020, detection rate for respiratory viruses was 28.15%, 2.15%, 12.8%, and 27.8% in winter, spring, summer, and autumn, respectively. While RSV detection rate in 2020 was 7.11%, 1.07%, 11%, and 22% in winter, spring, summer, and autumn, respectively. Interestingly, RSV detection rate in autumn was significantly higher in 2020 than autumn 2019 (22% > 2%, p < 0.05). Finally, detection rate for respiratory viruses was 6.7%, 4.14%, 9.4%, and 2.6% in winter, spring, summer, and autumn in 2021, respectively. While RSV detection rate in 2021 was 0.84%, 2.7%, 9.4%, and 19.3%in winter, spring, summer, and autumn, respectively. Similar to 2020, RSV detection rate in autumn 2021 was significantly higher (19.3%, p < 0.05) than other seasons of 2021.
Sequence alignments and phylogenetic analysis: All PCR amplicons (n = 93) were successfully sequenced and yielded good quality sequences for further analysis. All sequences were cleaned, edited, and aligned with representative reference genotypes. Among 93 RSV sequences, 56 (60.2%) and 37 (39.8%) were categorized into RSV-A and RSV-B respectively. There was no coinfection of RSV-A and RSV-B in tested samples. Sequence and phylogenetic analysis of G gene showed that RSV-A strains (56, 60.2%) belonged to ON1 genotype with the mean nucleotide sequence homology of 99%. They all aligned closely to a novel ON1 genotype which was first identified in 2010 in Ontario, Canada (Fig. 4). There was 96.4% sequence homology nucleotide level and 93.6% sequence homology at amino acid level between our study RSV-A strain (ON1) sequence and ON1 reference strain (Canada strain, ON67–1210, GenBank number: JN257693).All 56 RSV ON1 genotype were closely related to ON1 strain (MW455132.1, MN007037.1) previously isolated in Guangdong province ( percent identity 98–99%).
All RSV-B strains (n = 37, 39.8%) belonged to BA genotype and clustered with strains that were previously assigned to the BA-9 genotype with a 60-nucleotide duplication as shown in Fig. 5. A sequence homology of 94.8–97.2% at the nucleotide level and 89–94% at the amino acid level was observed among the sequences of RSV-B and the BA reference strain (AY333364) was. All 37 RSV-B strains of our study belonged to the BA9 genotype and were close to Guangzhou strains previously reported in China (MW527906). These findings indicate that ON1 and BA9 are the dominant genotypes in Guangzhou from 2017–2021.
Deduced Amino Acid Sequence Analysis:
We aligned and compared RSV-A strains (ON1 genotype) of our study with reference ON1 strain originated from Canada (Canada strain, ON67–1210, GenBank number: JN257693) and RSV-A prototype strain A2. An insertion of 24 amino acid (72 nucleotides) lengthened the G protein when compared to RSV-A prototype strain A2. We identified several amino acid substitutions at MLD1 and MLD2 of G protein when compared to the prototype ON1 strain (N257693). Five amino acid substitutions (T113I, V131D, N178 G, H258Q and H266 L) were notable in most of ON1 strains belonging to cluster 1 (ON1-5 clade). Two substitutions (N178G, H266L) were considered significantly important as they were present close to CX3C motif in CCD (N178G) and within antigenic site (265-273aa) indicating continuous evolution and adaptation of ON1 lineages in China which may play a role in RSV transmission and disease severity. A total of 228 amino acid changes were noticed at 43 different sites in the 56 sequences of second variable region of G protein (aa210 to aa321) of genotype ON1 compared with the reference ON1 genotype of RSV-A. The most amino acid variations were shown by two RSV-A strains of this study (GZ/2018/12157, GZ/2018/12256) which indicated variation in twenty-one and twenty amino acids respectively when compared to reference strain. Out of 43 sites, only 7 sites showed more than 2% frequency of amino acid change [V225A (3.5%), P230T (2.6%), T245A (4.4%), H258Q (18%), H266L (19%), L274P (3%), T320A (3%)]. While other 36 sites showed less than 1% frequency of amino acid change. Amino acid position 258 (H258Q, 21.31%), and 266 (H266L, 22.33%) showed the highest amino acid changes (Fig. 6). The analysis of potential N-glycosylation sites revealed two potential N glycosylation sites at amino acid positions 237 and 318 in second variable region of G protein of ON1 strains. However, N glycosylation was lost in two ON1 strains (GZ/2019/90411, GZ/2019/90557) due to substitutions at N237K and T239S, respectively.
Similarly, eight ON1strains (GZ/2018/11894, GZ/2019/82, GZ/2019/316, GZ/2019/365, GZ/2018/11496, GZ/2018/11467, GZ/2018/12157, GZ/2018/12526) lost N glycosylation due to T320A substitutions. Analysis of the second variable region predicted different pattern of O-glycosylation sites with 35–46 potential O-glycosylation sites among genotype ON1 sequences when compared to the Ontario reference strain. The amino acid positions most likely to have O-glycosylation in ON1 strains of present study are T211, T219, T220, T220, P222, L226, T227, T228, T231, P234, T235, T238, T239, T241, R244, T245, T246, T249, S250, T252, K253, T259, S260, T264, S267, T268, T269, S270, S275, P276, S277, T281, T282, S283, S299, S301, T305, T306, S307, S311, S313, S315, S316, S317, T319, T320 (refer to GZ/2018/11746).
We also aligned and compared second variable region of protein G of RSV-B strains (BA9 genotype) of our study with their respective reference sequences BA (AY333364). There were 396 amino acid substitutions in all 37 sequences of second variable region of G protein of RSV-B genotype (BA9), compared with BA reference sequences (AY333364). The amino acid variations analyses revealed 10 sites [K218T (9.3%), L223P (9.3%), S247P (9.3%), T254I (7.6%), T270I (9.3%), V271A (6.8%), T281I (9.3%), H287Y (8.8%), T290I (8.6%), T312I (8.3%)] with more than 5% frequency in amino acid changes and 102 sites with less than 5% frequency of amino acid changes in second variable region of G protein ( aa200 to aa312) (Fig. 7).
Interestingly, we observed several difference in amino acid frequency at position 254, 275, 290. 302 and 312 when compared reference strain of BA9 from China (KT765097, VRL-2016). Deduced amino acid analysis revealed two N-glycosylation sites at 296 and 310 amino acid positions of genotype BA9 sequences. However, these N-glycosylation sites at position 310 were missing in most of the BA9 strains due to due toT312I substitution and only four strains (GZ/2017/10135,GZ/208/11520,GZ/2018/12142,GZ/202/90370) showed potential N-glycosylation sites at residue 310. Similarly, three BA9 strains (GZ/2020/90484, GZ/2020/90510, GZ/2019/91194) lost potential N-glycosylation sites due to N296Y substitution. Analysis of the second variable region predicted different pattern of O-glycosylation with 27–50 potential O-glycosylation sites among genotype BA9 sequences when compared to the BA9 reference strain (KT765097). The amino acid positions most likely to have O-glycosylation in BA9 strains of present study are T211, P216, K218, T218, T222, T227, T228, I229, P231, T232, T236, T239, T240, R242, T244, S245, T246, S249, T250, T254, T255, T256, S257, T260,T264, S265, T266, S267, S269, T274,T275, T276, S277, T280, I281, S285, S288, T289, T290, T294, S297, T298, T300, T302, S304,S307,T308, S309, S311 and T312 for genotype BA9 (refer to GZ/2020/90597).