Hospitalizations due to acute hRSV lower respiratory tract infections during the 2023–2024 season showed that infants under five years of age were the primary affected demographic group.
hRSV-A exhibited a higher prevalence compared to hRS-B in both pediatric and adult patients. Particularly, infections by hRSV-A outnumbered those by hRSV-B by 3.4-fold in pediatric patients and approximately 1.7-fold in adults (Tables 1 and 2). Previous studies have documented that hRSV-A replicates faster and at higher titers (up to 2 log10 PFU/ml) than hRSV-B in airway human epithelial cell cultures21,22. However, similar viral loads have been found in respiratory samples from infants infected with either subgroup A or B, as measured by RT-qPCR or viral titration23,24, suggesting that factors besides replication rate contribute to the higher prevalence of hRSV-A.
Phylogenetic analysis demostrated that the prevalent lineages of hRSV-A were A.D.1.5, A.D.1.8, A.D.3, and A.D.5.2, while for hRSV-B, the predominant lineage was B.D.E.1. Studies conducted in three different countries reported the prevalence of the following clades after season 2021–2022: A.D.3, A.D.5, A.D.5.2 and B.D.E.1 in USA (Minnesota); A.D.3, A.D.5.2, B.D.4.1.1 and B.D.E.1 in China (Beijing); and A.D.1, A.D.3.1, A.D.5.2 and B.D.E.1 in Italy (Sicily)25,26,27. Accordingly, A.D.3, A.D.5.2 and B.D.E.1 may have global distribution27. Notably, the study from Beijing, China reported that B.D.E.1 became prevalent in November 2023 and affected a significantly higher proportion of patients ≥ 60 years (3-fold increase) than did its parent lineage B.D.4.1. However, B.D.E.1 was mostly associated with development of upper respiratory tract infections and non-severe pneumonia, conversely to the severe diseases caused by the previously dominant lineages A.D, A.D.3 and B.D.4.127.
Amino acid substitutions per viral genome and the substitution ratio (hRSV-A/hRSV-B) for each protein indicated higher variability in hRSV-A compared to hRSV-B. Total substitutions showed a tendency to increase in the hRSV-A isolates infecting the 0–16- year-old age group than in the 21–59-year-old group. This difference may be at least partially explained by the prevailing circulation of the hRSV-A in children, a host population with a naïve or minimally experienced immune system. This condition facilitates higher viral replication, which in turn can lead to an increased mutation rate28,29. The absence of RSV-B cases in the 12–45 age range (Fig. 3A and B) may reflect the limited number of total hRSV-B samples or potentially a lower susceptibility to hRSV-B infection in this age group, associated with asymptomatic or mild infections that did not require medical attention or sample collection.
As expected, we observed a high diversity of substitutions in G, F and L proteins. However, all substitutions in the F protein of hRSV-A showed frequencies < 40%, whereas many substitutions in G and L proteins displayed frequencies of up to100%.
The F protein contains six potential N-glycosylation sites located at positions N27, N70, N116, N120, N126 and N500. N116 and N126 are within the p27 fragment, which is cleavaged by furin-like cellular proteases30, while N120 is not conserved among different hRSV isolates31. Both, the double mutant N27Q/N70Q and the individual N500Q impair membrane fusion31. We did not identify substitutions in the N-glycosylation sites of our viral isolates. However, the substitution T122A (with a frequency of 28.8%), located in the consensus N-glycosylation sequence N-X-T/S was observed. It has been suggested that T122A might reduce the glycosylation at N120, although this likely has no significant impact on fusion activity31.
On the other hand, substitutions were detected in the F protein located within the antigenic sites ø and II. Particularly, the low frequency substitutions L258I was found in the Palivizumab binding site (residues 258 to 275) of hRSV-A. Zhu et al., reported 46 polymorphic sites in the extracellular region of the F protein of clinical isolates from children without prophylactic treatment32. Using microneutralization assays, it was determined that only the substitutions N262D and S275F conferred resistance to Palivizumab. These substitutions were identified in 2 of 145 hRSV A isolates by the authors and were considered as natural polymorphisms. In this study, changes in these positions were not detected.
Palivizumab resistance has been observed in 5–10% of immunocompromised infants with long-term infections and under treatment with Palivizumab. In such cases, substitutions N262D, K272E and S276N have been reported, although only the first two were associated with Palivizumab resistance32,33.
The recently approved prophylactic monoclonal antibody Nirsevimab targets a prefusion discontinuous neutralizing epitope within site Ø, spanning residues 62 to 69 and 196 to 21234. Although most residues within the binding site are conserved at a frequency of > 99%, a partial reduction in susceptibility to Nirsevimab has been associated with specific substitutions. These include N208D/S, K65Q/T and the dual substitution N67I/N208Y in hRSV-A, or dual substitutions K68N/N201S and K65Q/S211N in hRSV-B35,36.
In our study, we identified the substitution I206A (5.1%) in hRSV-A, and the substitutions M206I (3.8%), R209Q (7.7%), and S211N (96.2%) in hRSV-B. These substitutions within the Nirsevimab binding site have been documented as natural polymorphisms, and through microneutralization assays it has been determined that changes in the position 206 of the hRSV A-F protein do not modify the neutralization activity of Nirsevimab, while single mutations in positions 206 and 209 only partially reduce neutralization activity against hRSV-B isolates36. Given that the newly approved vaccines also target the prefusion conformation of the F protein, the variability we observed in antigenic sites, including low-frequency substitutions within sites Ø, raises the need to monitor for vaccine escape variants in Mexico where vaccines were approved in December 202437.
Other substitutions in F protein, localized in the signal peptide and p27 showed frequencies > 13%. Consistent with the observations of this study, it has been previously reported that domains with the greatest number of non-synonymous changes and amino acid positions with higher entropy values are within the signal peptide, p27, heptad repeat domain 2, antigenic site ø, and the transmembrane domain38.
As predicted, the G glycoprotein showed high variability, primarily in the second mucin-like region. However, the substitutions F168Y, S177G and N178G in the central conserved domain (CCD) of hRSV-A and substitution T198I in hRSV-B were also identified. Attachment of hRSV to primary respiratory epithelial cells occurs by binding of the G protein to CX3CR1 and infection is attenuated in viruses lacking the G protein or with mutations in the CX3C motif39. Although the highly glycosylated domains of the G protein are poor immunogens, antibodies against epitopes within the CCD have been detected. These antibodies can induce antibody-dependent cellular cytotoxicity or block the CX3C–CX3CR1 interaction40,41. Furthermore, substitutions as 177Q and 177R in the CCD enhance G protein immunogenicity and induce IgG antibodies that inhibit the CX3C–CX3CR1 interaction, thereby reducing pulmonary cell infiltration and lung damage in mice42. In this study, the substitution S177G at a low frequency (3.7%) in hRSV-A isolates were identified. It would be of interest to evaluate if the serum from individuals infected with this variant can reduce hRSV infection of primary airway epithelial cells.
On the other hand, Li et al., previously reported that substitutions T113I, V131D, N178 G, H258Q and H266 L in the G protein are associated with decreased disease severity in hospitalized infants43. We found the same five substitutions in hRSV-A isolates, with frequencies ranging from 6.8 to 10.2%. Further studies are necessary to assess the impact of these substitutions on viral infectivity.
Regarding the L protein or RNA-dependent RNA polymerase, we identified 190 and 63 different non-synonymous substitutions in hRSV-A and hRSV-B isolates, respectively. This enzyme is multifunctional, as it not only participates in the transcription and replication of the viral genome but also exhibits polyribonucleotidyltransferase (PRNTase) activity to add the cap structure, as well as methylase activity to methylate the cap44. Certain mutations in the L protein have been identified in the context of studies with antiviral drugs that inhibit its enzymatic activity45,46,47. Substitutions associated with antiviral drug resistance were not identified in this study. However, substitutions R1339P and G1855R within the catalytic pocket of the capping domain and the SAM/SAH GxGxGx binding motif of the MTase domain, respectively were observed20. Both substitutions were present at a frequency of 1.7%.
Substitutions in hRSV proteins other than F, G and L have been less extensively studied. Nevertheless, the growing availability of complete hRSV genome sequences will facilitate the surveillance of specific substitutions and their frequencies, thereby contributing to a better understanding of viral evolution and the effectiveness of prevention and treatment strategies.
This study represents the first detailed genomic characterization of hRSV-A and hRSV-B isolates from hospitalized pediatric and adult patients in Mexico City during the 2023–2024 season, contributing to global surveillance efforts. These results confirm the previously described predominance of hRSV-A and its high variability in the G protein, but also show demonstrate the variability in the less-studied P, M2-1 and M2-2 proteins. The local circulation of globally distributed lineages such as A.D.5.2 and B.D.E.1 was observed, as well as the occurrence of substitutions within the antigenic sites Ø and II of the F protein. These mutations speak to the need for continuous monitoring to assess potential impacts on vaccine and monoclonal antibody effectiveness in our geographic region. Although age-stratified analysis did not reveal significant differences in substitution frequency, we observed a tendency toward a higher number of substitutions in viral genomes isolated from infants, which might reflect age-related differences in immune pressure. However, further studies with higher number of samples are necessary to validate this hypothesis.
The limitations of our study include that samples were mainly collected from hospitalized patients at the INER, located in Mexico City. This restricts the assessment of hRSV lineages circulation in other regions of Mexico and excludes the genetic characterization of viruses infecting individuals with mild disease. Also, the method used to sequence the complete viral genome excluded samples with low viral loads (Ct > 25) preventing the evaluation of potential associations between viral load and hRSV lineages. Finally, while substitutions previously related to resistance to monoclonal antibodies were identified, this study did not include functional assays to confirm their effect on the efficacy of preventive treatments such as Palivizumab or Nirsevimab.
