Association of SNP-SNP Interactions of Surfactant Protein Genes with Pediatric Acute Respiratory Failure

Abstract

The hallmarks of pediatric acute respiratory failure (ARF) are dysregulated inflammation and surfactant dysfunction. The objective is to study association of surfactant protein (SP) genes' single nucleotide polymorphisms (SNPs) with ARF and its morbidity: pulmonary dysfunction at discharge (PDAD), employing a single-, two-, and three-SNP interaction model. We enrolled 468 newborn controls and 248 children aged ≤ 24 months with ARF; 86 developed PDAD. Using quantitative genetic principles, we tested the association of SP genes SNPs with ARF and PDAD. We observed a dominant effect of rs4715 of the SFTPC on ARF risk. In a three-SNP model, we found (a) 34 significant interactions among SNPs of SFTPA1, SFTPA2, and SFTPC associated with ARF (p = 0.000000002-0.05); 15 and 19 of those interactions were associated with increased and decreased risk for ARF, respectively; (b) intergenic SNP-SNP interactions of both hydrophobic and hydrophilic SP genes associated with PDAD (p = 0.00002-0.03). The majority of intra- and intergenic interactions associated with ARF involve the SFTPA2 SNPs, whereas most of the intra- and intergenic interactions associated with PDAD are of SFTPA1 SNPs. We also observed a dominant effect of haplotypes GG of SFTPA1 associated with increased and AA of SFTPC associated with decreased ARF risk (p = 0.02). To the best of our knowledge, this is the first study showing an association of complex interactions of SP genes with ARF and PDAD. Our data indicate that SP genes polymorphisms may contribute to ARF pathogenesis and subsequent PDAD and/or may serve as markers for disease susceptibility in healthy children.

Keywords: Pediatric acute respiratory failure; SFTPB; SFTPC; SFTPD; SNP–SNP interaction; Surfactant protein gene A1 (SFTPA1), SFTPA2.

Figure 1

Study outline. Previously healthy children with acute respiratory failure (ARF) are enrolled in the study (n = 250). Two children subsequently diagnosed with cystic fibrosis were excluded from final analysis (n = 248). The statistical analysis (shown by a blue double headed arrow) is done between the cases (ARF, n = 248) and controls (healthy newborns, n = 468). Then, further analysis was performed between the cohort of children with and without pulmonary dysfunction at discharge (PDAD) shown in red double headed arrow (n = 86 vs 162).

Figure 2

(A) shows schematic presentation of SFTPA1 and (B) shows SFTPC gene and transmission of haplotypes in ARF. The relative location of genes is shown from centromere (C) to telomere (T). In each panel, the number above the solid black line indicates the amino acid number and corresponding nucleotide change shown in parenthesis. The SNP id is shown below the black line. The arrows indicate transcriptional orientation. The transmitted haplotypes and corresponding amino acid changes are shown in bold in a two-SNP model. A shows the location of the SFTPA1 gene in blue on the long arm of chromosome 10. Haplotype “GG” shown in black line, constituted by rs1136450 Val(G) at codon 50 and rs1136451 Pro(G) at codon 62 within the SFTPA1 gene, are associated with increased ARF risk in a previously healthy children, p = 0.02, OR = 9.3 (1.7–49.6). B shows the SFTPC gene in orange and its orientation in terms of the centromere and telomere. The haplotype “AA”, shown in red line, constituted by rs4715 Asn(A) at codon 138 and rs1124 Asn(A) at codon 186 within the SFTPC gene, is shown to associate with lower risk for ARF, p = 0.02, OR = 0.4 (0.2–0.7). Both significant haplotypes have a dominant effect on ARF. Of note, the physical location of the transmitted SNPs in these haplotypes is very close, as shown in Figure.

Figure 3

Common association of SNP–SNP interactions with ARF and PDAD. Figure shows SNP–SNP interactions (black and red double head arrows) found in common in two sets of comparisons. In one, comparison between pediatric acute respiratory failure (ARF) to newborn controls; an odds ratio (OR) of 0.5 indicates that these interactions associate with lower risk for ARF. In the other, comparison between those with pulmonary dysfunction at discharge (PDAD) subgroup compared to those who did not develop PDAD after adjusting for age, positive bacterial culture, and number of ventilator days, an OR of 2 to 2.5 indicates that these interactions associate with increased susceptibility of ARF survivors to develop PDAD. The SNP of the surfactant gene SFTPC, encoding the hydrophobic SP-C, is depicted within the orange rectangle on the right side; on the left, SNPs of the surfactant genes, SFTPA1 and SFTPA2, encoding the hydrophilic surfactant proteins SP-A1 and SP-A2, are depicted within the light- and dark-blue rectangle, respectively. Arrows represent the significant interactions in a three-SNPs model. These interactions are intergenic and dominant × dominant × dominant (d1 × d2 × d3) among the SNPs of SFTPA1, SFTPA2, and SFTPC. Red arrows represent the d1 × d2 × d3 interactions among two SNPs of SFTPA1 and one SNP of SFTPC (rs1136451 × rs4253527 × rs4715, p = 0.00002–0.008), while black arrows represent three SNP interactions among the SNPs of SFTPA2, SFTPA1, and SFTPC (rs17886395 × rs4253527 × rs4715, p = 0.0001–0.005). Of note, rs4253527 (SFTPA1) and rs4715 (SFTPC) are common in both interactions and represent both hydrophilic and hydrophobic surfactant proteins.

Figure 4

Association of SNP–SNP interactions of SP genes with PDAD. (A) shows interaction of SNPs of SFTPA1, SFTPB, and SFTPC genes and (B) shows interaction of SNPs of hydrophilic SPs (SFTPA1, SFTPA2, and SFTPD) alone. SP gene SNP–SNP interactions associated with PDAD compared to those who did not develop PDAD after adjusting for age, positive bacterial culture, and ventilator days. All interactions are intergenic and in a dominant form. In A, on the right, one SNP of the surfactant gene SFTPB and one SNP of the SFTPC gene, encoding the hydrophobic SP-B and SP-C, respectively, is shown within the dark and light orange rectangle, respectively. The SFTPA1 gene is shown in light blue on the left side. Solid arrows represent interactions between the hydrophilic SFTPA1 gene, and the hydrophobic SFTPB and SFTPC genes (d1 × d2 × d3 = rs1059047 × rs1136450 × rs1130866, p = 0.003–0.02, rs1059047 × rs1136451 × rs4715, p = 0.003–0.03). The odds ratio (OR) of less than one (0.4–0.6) indicates that both of these interactions associate with lower risk for PDAD. In B, the SFTPD gene is shown in purple on the right side. SFTPA1 and SFTPA2 genes are shown in light and dark blue, respectively, on the left side. Dotted arrows represent interactions among SNPs of the hydrophilic surfactant proteins, SFTPA2, SFTPA1, and SFTPD (rs1059046 × rs1136450 × rs2243639, p = 0.005–0.02, and rs1059046 × rs721917 × rs2243639, p = 0.00005–0.01, OR = 2.3). The OR of rs1059046 × rs1136450 × rs2243639 interaction is less than 1 (0.5–0.6), indicating that this interaction associate with lower risk for PDAD. In contrast, the OR of rs1059046 × rs721917 × rs2243639 interaction is more than 1 (1.9–2.3) indicating that this interaction is associated with increased susceptibility for PDAD.