Hydrophilic But Not Hydrophobic Surfactant Protein Genetic Variants Are Associated With Severe Acute Respiratory Syncytial Virus Infection in Children

Abstract

Respiratory syncytial virus (RSV) is the leading cause of lower respiratory tract infection-related hospitalization in the first year of life. Surfactant dysfunction is central to pathophysiologic mechanisms of various pulmonary diseases including RSV. We hypothesized that RSV severity is associated with single nucleotide polymorphisms (SNPs) of surfactant proteins (SPs). We prospectively enrolled 405 RSV-positive children and divided them into moderate and severe RSV disease. DNA was extracted and genotyped for sixteen specific SP gene SNPs. SP-A1 and A2 haplotypes were assigned. The association of RSV severity with SP gene SNPs was investigated by multivariate logistic regression. A likelihood ratio test was used to test the goodness of fit between two models (one with clinical and demographic data alone and another that included genetic variants). p ≤ 0.05 denotes statistical significance. A molecular dynamics simulation was done to determine the impact of the SFTPA2 rs1965708 on the SP-A behavior under various conditions. Infants with severe disease were more likely to be younger, of lower weight, and exposed to household pets and smoking, as well as having co-infection on admission. A decreased risk of severe RSV was associated with the rs17886395_C of the SFTPA2 and rs2243639_A of the SFTPD, whereas an increased risk was associated with the rs1059047_C of the SFTPA1. RSV severity was not associated with SNPs of SFTPB and SFTPC. An increased risk of severe RSV was associated with the 1A⁰ genotype of SFTPA2 in its homozygous or heterozygous form with 1A³. A molecular dynamic simulation study of SP-A variants that differ in amino acid 223, an important amino acid change (Q223K) between 1A⁰ and 1A³, showed no major impact on the behavior of these two variants except for higher thermodynamic stability of the K223 variant. The likelihood ratio test showed that the model with multi-allelic variants along with clinical and demographic data was a better fit to predict RSV severity. In summary, RSV severity was associated with hydrophilic (but not with hydrophobic) SPs gene variants. Collectively, our findings show that SP gene variants may play a key role in RSV infection and have a potential role in prognostication.

Keywords: SFTPA1; SFTPA2; SFTPB; SFTPC; SFTPD; respiratory sycytial virus; single nucelotide polymorphisms; surfacant protein genetic variant.

Figure 1

Forest plot represents predictors of severe RSV using mutivariable logistic regression analysis. X-axis and Y-axis represents odds ratio (OR) and variables, respectively. Dotted line represents OR of 1. Based on OR, 1A⁰/1A⁰, 1A⁰/1A³, pet exposure and co-infection are associated with increase risk of severe RSV (OR > 1), whereas, older age is associated with decreased risk of severe RSV (OR < 1).

Figure 2

Root-mean-square fluctuation (RMSF) in the backbone and sidechain of the Q223 and K223 SP-A variants. SP-A homo-trimers consisting of three identical monomers of SP-A1 or SP-A2 that included amino acids 104-224 were studied. The RMSF trimer data of each of the two variants (Q223 and K223) were analyzed for each of the 3 identical monomers present in each trimer to study the behavior of SP-A trimer. (A–C) represent monomers 1, 2, and 3, respectively, for Q223 trimer, and (D–F) represent monomers 1,2,3, respectively, for K223 trimer. The amino acid residue number is on the x-axis and RMSF is on the y-axis. The backbone and sidechain of the Q223 variant, are depicted in orange and gray colors, respectively. Whereas, blue and green colors depict the backbone and sidechain of the K223 SP-A, respectively. The position of peaks did not differ significantly between Q223 and K223 variants, however, the peak of each residue, in general, is greater for Q223 compared to the K223. The region around amino acid 223 (the site of variation), has more intense peaks, indicating a region with great atomic movement.