Abstract
Objective
To determine whether synonymous variants in the adenosine triphosphate-binding cassette A3 transporter (ABCA3) gene increase the risk for neonatal respiratory distress syndrome (RDS) in term and late preterm infants of European and African descent.
Study design
Using next-generation pooled sequencing of race-stratified DNA samples from infants of European and African descent at $34 weeks gestation with and without RDS (n = 503), we scanned all exons of ABCA3, validated each synonymous variant with an independent genotyping platform, and evaluated race-stratified disease risk associated with common synonymous variants and collapsed frequencies of rare synonymous variants.
Results
The synonymous ABCA3 variant frequency spectrum differs between infants of European descent and those of African descent. Using in silico prediction programs and statistical strategies, we found no potentially disruptive synonymous ABCA3 variants or evidence of selection pressure. Individual common synonymous variants and collapsed frequencies of rare synonymous variants did not increase disease risk in term and late-preterm infants of European or African descent.
Conclusion
In contrast to rare, nonsynonymous ABCA3 mutations, synonymous ABCA3 variants do not increase the risk for neonatal RDS among term and late-preterm infants of European or African descent.
Neonatal respiratory distress syndrome (RDS) results from insufficiency of pulmonary surfactant, a phospholipid–protein complex that is synthesized, packaged, and exocytosed by alveolar type 2 cells, decreases surface tension, and maintains alveolar expansion at end expiration.1 RDS is generally attributed to developmental insufficiency of pulmonary surfactant production; however, genetic mechanisms also contribute to the risk for neonatal RDS.2–7
Adenosine triphosphate-binding cassette A3 transporter (ABCA3) is a member of the highly conserved family of adenosine triphosphate binding cassette transporters that bind and hydrolyze adenosine triphosphate to transport substrates across cellular membranes.8 ABCA3 is most highly expressed in the lung and is localized to the limiting membranes of lamellar bodies, intracellular storage organelles of pulmonary surfactant.9,10 Rare, recessive, nonsynonymous mutations in ABCA3 are associated with lethal neonatal RDS and chronic respiratory disease in children.5,11 Recently, single, rare, nonsynonymous mutations in ABCA3 were associated with reversible RDS in term and late-preterm infants of European descent.7
Although nonsynonymous ABCA3 mutations that change the amino acids coded into that protein are known to increase the risk of neonatal RDS,5,7,12 much less is known about synonymous variants that do not change the amino acid sequence but may alter intron–exon splicing, splicing control elements, messenger RNA stability, translation efficiency, or protein folding.13–18 Two synonymous ABCA3 variants have been associated with the risk of neonatal RDS.19,20 The synonymous variant p.F353F, which resides in the transmembrane domain, was associated with a prolonged course of RDS in preterm Finnish infants,19 and p.P585P, which resides in the nucleotide binding domain, was overrepresented in preterm Chinese infants with RDS.20
Given that mutations in ABCA3 can cause severe neonatal RDS, the evaluation of term and late-preterm infants with progressive respiratory failure unresponsive to medical management frequently includes ABCA3 sequencing to establish a diagnosis of ABCA3 deficiency.21 Because most ABCA3 mutations are rare, private, and have not been evaluated in surrogate cell systems,22,23 clinicians must rely on results of in silico prediction algorithms24–26 and the opinions of experts. Even though synonymous variants are frequently identified with such genetic sequencing, prognostic information for these variants is limited. Thus, using high-resolution, high-throughput, next-generation exonic sequencing; computational algorithms for variant discovery; in silico programs to predict functionality; independent validation of variants; and statistical strategies to compare common synonymous variant and collapsed rare synonymous variant frequencies, we examined the associations of synonymous ABCA3 variants with the risk of neonatal RDS in term and late-preterm infants of European and African descent.
Methods
We used DNA collected from a previously reported prospectively enrolled cohort of newborn infants with and without RDS, ≥34 weeks gestational age, and maternally designated European or African descent recruited from the nurseries at Washington University Medical Center7 (Table I). We defined RDS as a requirement for supplemental oxygen (fraction of inspired oxygen ≥0.3), chest radiograph findings consistent with RDS, and the need for continuous positive airway pressure or mechanical ventilation within the first 48 hours of life.6,7 Infants without RDS (non-RDS group) had no respiratory symptoms and were hospitalized for other neonatal problems. We assigned gestational age based on the best obstetrical estimate, and we excluded infants with cardiopulmonary malformations, pulmonary hypoplasia, culture-positive sepsis, chromosomal anomalies, known surfactant mutations, or rapidly resolving RDS (within <24 hours of birth). We randomly excluded 1 of each set of monozygotic twins (n = 3) and twins in whom zygosity could not be reliably determined (n = 2). We extracted details of the respiratory course and outcome from the clinical chart. This study was reviewed and approved by the Washington University School of Medicine’s Human Research Protection Office.
Table I.
Characteristics of European and African descent disease-based groups (n = 503)
| European descent | African descent | |||||
|---|---|---|---|---|---|---|
| RDS (n = 109) | Non-RDS (n = 158) | P | RDS (n = 44) | Non-RDS (n = 192) | P | |
| Sex, n (%) | ||||||
| Female | 45 (41) | 72 (46) | .49 | 10 (0.23) | 99 (0.52) | <.001 |
| Male | 64 (59) | 86 (54) | 34 (0.77) | 93 (0.48) | ||
| Gestational age, wk, mean ± SD | 36.9 ± 1.7 | 38.2 ± 1.6 | <.001 | 37.6 ± 2.7 | 38.9 ± 1.7 | .003 |
| Birth weight, kg, mean ± SD | 3.1 ± 0.6 | 3.1 ± 0.7 | .50 | 2.9 ± 1.0 | 3.1 ± 0.4 | .07 |
| Route of delivery, n (%) | ||||||
| Vaginal | 54 (0.49) | 71 (0.45) | .53 | 15 (0.34) | 131 (0.68) | <.001 |
| Cesarean | 55 (0.51) | 87 (0.55) | 29 (0.66) | 61 (0.32) | ||
DNA Isolation and Pool Preparation
We isolated DNA from blood samples using Puregene DNA isolation kits (Qiagen, Valencia, California)6,7 and combined equimolar amounts from each individual into 4 race-stratified pools: infants of African descent with RDS (n = 44), infants of African descent without RDS (n = 196), infants of European descent with RDS (n = 112), and infants of European descent without RDS (n = 161).
Next-Generation Sequencing and Validation
We used an Illumina next-generation sequencing platform to sequence all exons and flanking regions (approximately 50 base pairs) of ABCA3 (data available on request).27 To optimize the selection of significance thresholds for detection of rare variants in each sequencing run, we added a 1934-bp oligonucleotide without variation and a 335-bp oligonucleotide containing 15 known insertions, deletions, and substitutions at a frequency of <1 allele per pool.28 Inclusion of negative and positive controls allowed run-specific error models to achieve high sensitivity (0.99) and specificity (0.99) for detecting rare variants within each pool. We sequenced approximately 37 kb per individual, with a mean coverage of 82×.
We then used the computational algorithm SPLINTER28 to detect rare (ie, minor allele frequency [MAF] <0.01) and common (MAF ≥0.01) synonymous variants. Each variant was confirmed with an independent genotyping strategy (Sequenom, TaqMan, or Sanger resequencing) and linked to its individual sample (data available on request).We had insufficient DNA for 10 infants (4 of African descent and 6 of European descent) to complete the validation studies, and thus we excluded these infants from all further analyses (Table I).
In Silico Prediction of Functionality
We used Alamut 2.3 (Interactive Biosoftware, Rouen, France), which combines the results of 7 splicing prediction algorithms— Human Splicing Finder (www.umd.be/HSF),29 GeneSplicer (http://www.cbcb.umd.edu/software/GeneSplicer),30 MaxEntScan (genes.mit.edu/burgelab/maxent/Xmaxentscan_scoreseq.html),31 NNSplice,32 Splice Site Finder-Like,33 ESE-Finder (http://rulai.cshl.edu/tools/ESE), and RESCUE-ESE (http://genes.mit.edu/burgelab/rescue-ese/)—to predict whether a synonymous variant would alter predicted intron–exon splicing patterns. We used the Genome Variation Server (http://gvs.gs.washington.edu/GVS137/index.jsp) to determine whether the common (MAF ≥0.01) synonymous variants were in linkage disequilibrium.
Statistical Analyses
We used χ2 and Fisher exact tests to determine whether common ABCA3 synonymous variants (MAF ≥0.01) were in Hardy-Weinberg equilibrium. We compared race-stratified frequencies of individual common (MAF ≥0.01) synonymous variants in infants with and without RDS using the χ2 test (Table II). We compared collapsed frequencies of rare (MAF <0.01) synonymous variants in infants with and without RDS using χ2 and Fisher exact probability tests (Table II).34 Because an individual is unlikely to carry more than 1 rare synonymous variant at a single gene locus, the number of rare variants in a single gene can be collapsed for statistical purposes and compared using a univariate test.34 We used logistic regression to determine whether the presence of a synonymous variant increases the risk of RDS or has a gestational age-specific interaction. We used a gene dosage model35 to determine whether carrying more than 1 synonymous variant imparts an additive risk for RDS. We used logistic regression to determine whether carrying a synonymous variant interacts with the presence of a single rare nonsynonymous ABCA3 mutation to increase the risk of RDS. We used the Fisher exact test and Student t test to compare demographic data and severity measures between groups. We linked these data with the previously reported nonsynonymous mutation data,7 and performed logistic regression to determine whether synonymous variants statistically interacted with rare nonsynonymous ABCA3 mutations to increase the risk of neonatal RDS. We used the Exome Variant Server (National Heart, Lung, and Blood Institute Exome Sequencing Project [ESP]; http://snp.gs.washington.edu/EVS/, release ESP6500MF, accessed March 2014), a database of more than 6500 individuals from up to 18 different US populations that participated in longitudinal cardiovascular and pulmonary-related research, to determine population-based frequencies of synonymous variants and compare these frequencies with those found in our cohorts.
Table II.
Synonymous variants identified among infants of European descent
| Variant | RDS (n = 109) | Non-RDS (n = 158) | P | ESP (n = 4300) | Distance from splice site, bp |
|---|---|---|---|---|---|
| V150V | 1 (0.0046) | 5 (0.016) | .41 | 6 (0.00070) | 3 |
| A227A | 5 (0.023) | 8 (0.025) | 1.0 | 192 (0.022) | 68 |
| F353F | 26 (0.12) | 41 (0.13) | .79 | 847 (0.098) | 53 |
| P585P | 40 (0.18) | 61 (0.19) | .82 | 1702 (0.20) | 14 |
| A928A | 0 | 1 (0.0032) | 1.0 | 0 | 84 |
| L982L | 1 (0.0046) | 0 | .41 | 1 (0.00012) | 59 |
| A1027A | 0 | 1 (0.0032) | 1.0 | 0 | 77 |
| S1372S | 18 (0.083) | 36 (0.11) | .25 | 729 (0.085) | 49 |
| E1618E | 0 | 1 (0.0032) | 1.0 | 0 | 56 |
| V1648V | 0 | 1 (0.0032) | 1.0 | 27 (0.0063) | 35 |
| P1653P | 0 | 2 (0.0063) | .52 | 0 | 25 |
| Total synonymous variant frequency | 91 (0.42)* | 157 (0.50)* | .078 | 3652 (0.42)*,† | |
| Collapsed rare (MAF <0.01) synonymous variant frequency‡ | 2 (0.018)§ | 11 (0.070)§ | .081 | 182 (0.042)†,§ |
Number (carrier allele frequency, all synonymous variants).
Includes additional synonymous variants detected only in the ESP, not in our cohort.
MAF based on ESP allele frequency results.
Number (carrier rare allele frequency, assuming 1 rare synonymous variant per individual).
Results
Infants of European Descent
We identified synonymous ABCA3 variants in 181 of the 267 infants, with 54 infants carrying more than 1 synonymous variant (Table III; available at www.jpeds.com). Four synonymous variants (p.A227A, p.F353F, p.P585P, and p.S1372S) were common, and none deviated from Hardy-Weinberg equilibrium or were in linkage disequilibrium (maximum R2 = 0.053; data available on request). Using Alamut, we found that none of the synonymous variants was predicted to disrupt intron–exon splicing.
Table III.
Number of patients with synonymous ABCA3 variants
| Group | 0 variants | 1 variant | 2 variants | 3 variants | 4 variants | Rare nonsynonymous ABCA3 mutation |
|---|---|---|---|---|---|---|
| European descent | 86 | 127 | 41 | 13 | 0 | 22 |
| African descent | 78 | 91 | 47 | 16 | 4 | 5 |
There were no differences in the individual common synonymous ABCA3 variant frequencies or the collapsed rare synonymous variant frequencies between infants of European descent with RDS and those without RDS (Table II). We used logistic regression models that included gestational age, sex, and mode of delivery, 3 covariates known to be associated with the risk of neonatal RDS, to determine that neither the presence of a single synonymous variant (P = .37) nor the number of synonymous variants (P = .26) was significantly predictive of RDS status. Although gestational age was an independent predictor of RDS among infants of European descent (P < .001), further modeling did not reveal a statistically significant interaction between the presence of a synonymous ABCA3 variant and gestational age (P = .15).
Even though the infants of European descent with RDS had a lower mean gestational age compared with those without RDS, there was no difference in mean gestational age (P = .069) or birth weight (P = .56) between infants with and without synonymous ABCA3 variants (Tables I and IV; Table IV available at www.jpeds.com). Moreover, we found no differences in measures of disease severity between infants of European descent with RDS with an ABCA3 mutation and those without an ABCA3 mutation, including duration of the need for mechanical ventilation, supplemental oxygen, pneumothorax, need for extracorporeal membrane oxygenation, need for home oxygen, and death (Table V; available at www.jpeds.com). The lack of statistically significant interactions between the presence of a synonymous variant (P = .80) or number of synonymous variants (P = .56) and rare nonsynonymous mutations in ABCA3 suggests that nonsynonymous mutations act independently to increase the risk of neonatal RDS in term and late-preterm infants of European descent.7
Table IV.
Clinical characteristics of infants of European descent with and without synonymous ABCA3 variants
| Characteristic | Variant present (n = 181) |
Variant absent (n = 86) |
P |
|---|---|---|---|
| Sex | |||
| Female | 80 (44) | 37 (43) | .86 |
| Male | 101 (56) | 49 (57) | |
| Gestational age, wk, mean SD | 37.8 ± 1.7 | 37.4 ± 1.7 | .069 |
| Birth weight, kg, mean SD | 3.1 ± 0.7 | 3.1 ± 0.6 | .56 |
| Route of delivery, | |||
| n (%) | |||
| Vaginal | 83 (0.46) | 42 (0.49) | .65 |
| Cesarean | 98 (0.54) | 44 (0.51) |
Table V.
Disease severity measurements among infants of European descent with RDS with and without synonymous ABCA3 variants
| Characteristic | Variant present (n = 68) |
Variant absent (n = 41) |
P |
|---|---|---|---|
| Duration of ventilation, d, mean SD | 11 ± 16 | 9 ± 11 | .36 |
| Duration of oxygen, d, mean SD | 26 ± 65 | 15 ± 14 | .17 |
| Pneumothorax, % | 25 | 10 | .18 |
| ECMO, % | 2 | 3 | .36 |
| Home oxygen therapy, % | 6 | 6 | .35 |
| Death, % | 2 | 4 | .20 |
ECMO, extracorporeal membrane oxygenation.
Infants of African Descent
We found synonymous variants in 158 of 236 infants of African descent, 67 of whom carried more than 1 synonymous variant (Table III). In addition to the 3 synonymous variants (p.F353F, p.P585P, and p.S1372S) commonly seen in infants of European descent, p.V150V and p.V623V were common in the infants of African descent as well (Table VI). None of these common variants deviated from Hardy-Weinberg equilibrium or were in linkage disequilibrium (maximum R2 = 0.024; data available on request). Using Alamut, none of the synonymous variants identified in our infants was predicted to disrupt intron–exon splicing.
Table VI.
Synonymous variants identified among infants of African descent
| Variant | RDS (n = 44) | Non-RDS (n = 192) | P | ESP (n = 2198) | Distance from splice site, bp |
|---|---|---|---|---|---|
| V150V | 13 (0.15) | 50 (0.13) | .61 | 498 (0.11) | 3 |
| A227A | 1 (0.011) | 1 (0.0026) | .33 | 11 (0.0025) | 68 |
| A279A | 1 (0.011) | 0 | .18 | 0 | 37 |
| I318I | 1 (0.011) | 1 (0.0026) | .33 | 20 (0.0045) | 37 |
| P343P | 0 | 1 (0.0026) | 1.0 | 6 (0.0014) | 39 |
| F353F | 4 (0.045) | 30 (0.077) | .37 | 383 (0.087) | 53 |
| P585P | 7 (0.080) | 33 (0.084) | 1.0 | 329 (0.075) | 14 |
| V623V | 1 (0.011) | 3 (0.0077) | .56 | 59 (0.013) | 28 |
| Y754Y | 0 | 2 (0.0051) | 1.0 | 0 | 2 |
| H780H | 1 (0.011) | 1 (0.0026) | .33 | 19 (0.0043) | 75 |
| S1372S | 21 (0.24) | 77 (0.20) | .38 | 968 (0.22) | 49 |
| Total synonymous variant frequency | 41 (0.47)* | 199 (0.51)* | .50 | 2404 (0.55)*,† | |
| Collapsed rare (MAF <0.01) synonymous variant frequency§ | 4 (0.091)‡ | 6 (0.031)‡ | .089 | 167 (0.076)†,‡ |
Number (carrier allele frequency, all synonymous variants).
Includes additional synonymous variants detected only in the ESP, not in our cohort.
Number (carrier rare synonymous variant frequency, assuming 1 rare variant per individual).
MAF based on ESP allele frequency results.
We found no differences in the individual common synonymous ABCA3 variant frequencies or the collapsed rare synonymous variant frequencies between infants of African descent with RDS and those without RDS (Table VI). We applied the same logistic regression models used in the European descent cohort and found that neither the presence of a single synonymous variant (P = .87) nor the number of synonymous variants (P = .47) was significantly predictive of RDS status. Similar to the European descent cohort, there was no statistically significant interaction between the presence of a synonymous ABCA3 variant and gestational age (P = .80), and we found no differences in disease severity between infants of African descent with RDS with synonymous ABCA3 variants and those without synonymous ABCA3 variants (Table VII; available at www.jpeds.com). We also found no statistically significant interactions between the presence (P = .75) or number (P = .78) of synonymous variants and rare nonsynonymous ABCA3 mutations to increase the risk of neonatal RDS.
Table VII.
Disease severity measurements among infants of African descent with RDS with and without ABCA3 synonymous variants
| Characteristic | Variant present (n = 30) |
Variant absent (n = 14) |
P |
|---|---|---|---|
| Duration of ventilation, d, mean ± SD | 11 ± 30 | 4 ± 5 | .29 |
| Duration of oxygen, d, mean ± SD | 17 ± 31 | 9 ± 8 | .21 |
| Pneumothorax, % | 4 | 0 | .29 |
| ECMO, % | 0 | 0 | NA |
| Home oxygen therapy, % | 2 | 0 | 1.0 |
| Death, % | 2 | 0 | 1.0 |
NA, not applicable.
Although the mean gestational age was lower in the infants of African descent with RDS compared with those without RDS, there was no difference in mean gestational age (P = .48) or birth weight (P = .44) between infants with synonymous variants and those without synonymous variants (data available on request). In addition, there were fewer female infants of African descent with RDS than males, likely related to our consecutive enrollment strategy and the lower incidence of RDS among females of African descent compared with males of similar gestational age.36 Infants of African descent with RDS were more likely to be delivered via cesarean delivery compared with those without RDS. Although transient tachypnea of the newborn is associated with cesarean delivery,37 we only included infants with respiratory symptoms persisting for >24 hours.
Using the Exome Variant Server, we found no difference in the frequencies of common synonymous ABCA3 variants between our cohort and individuals of European and African descent (Tables II, VIII, and IX; Tables VIII and IX available at www.jpeds.com). In addition, the collapsed frequency of rare synonymous variants in the ESP European descent cohort was similar to that in our infants of European descent without RDS (4.2% vs 7.0%; P = .098). More rare synonymous variants were present in the ESP African descent cohort compared with our infants of African descent without RDS (7.6% vs 3.1%; P = .022). This difference may reflect the greater number of individuals of African descent in the ESP, which enhances the ability to detect rare variants.38
Table VIII.
Synonymous variants among infants of European descent in the ESP
| Variant | ESP (n = 4300)* |
Distance from splice site, bp |
|---|---|---|
| T25T | 1 (0.00012) | 21 |
| L31L | 1 (0.00012) | 39 |
| P58P | 1 (0.00012) | 116 |
| T92T | 1 (0.00012) | 44 |
| A131A | 16 (0.0019) | 55 |
| V150V | 6 (0.00070) | 3 |
| Y158Y | 1 (0.00012) | 27 |
| T168T | 1 (0.00012) | 57 |
| P193P | 1 (0.00012) | 35 |
| A227A | 192 (0.022) | 68 |
| V240V | 1 (0.00012) | 107 |
| P248P | 1 (0.00012) | 130 |
| P249P | 6 (0.00070) | 127 |
| I251I | 5 (0.00058) | 121 |
| P254P | 2 (0.00023) | 112 |
| L268L | 1 (0.00012) | 70 |
| T277T | 1 (0.00012) | 43 |
| V282V | 3 (0.00035) | 28 |
| L317L | 1 (0.00012) | 40 |
| I318I | 1 (0.00012) | 37 |
| A319A | 1 (0.00012) | 34 |
| A336A | 3 (0.00035) | 18 |
| P343P | 1 (0.00012) | 39 |
| L351L | 2 (0.00023) | 61 |
| F353F | 847 (0.098) | 53 |
| F377F | 1 (0.00012) | 20 |
| P394P | 1 (0.00012) | 71 |
| A499A | 1 (0.00012) | 30 |
| A528A | 1 (0.00012) | 28 |
| T574T | 1 (0.00012) | 20 |
| P585P | 1702 (0.20) | 14 |
| T586T | 1 (0.00012) | 17 |
| S593S | 1 (0.00012) | 38 |
| D600D | 1 (0.00012) | 59 |
| V602V | 1 (0.00012) | 65 |
| R663R | 3 (0.00035) | 64 |
| L679L | 1 (0.00012) | 16 |
| D724D | 2 (0.00023) | 92 |
| I733I | 8 (0.00093) | 65 |
| Y754Y | 1 (0.00012) | 2 |
| A756A | 17 (0.0020) | 5 |
| H780H | 1 (0.00012) | 75 |
| V839V | 1 (0.00012) | 4 |
| D882D | 3 (0.00035) | 55 |
| D952D | 6 (0.00070) | 149 |
| L982L | 1 (0.00012) | 59 |
| D988D | 2 (0.00023) | 41 |
| A1055A | 1 (0.00012) | 114 |
| H1069H | 2 (0.00023) | 72 |
| F1077F | 7 (0.00081) | 48 |
| P1078P | 1 (0.00012) | 45 |
| E1093E | 6 (0.00070) | 1 |
| N1103N | 1 (0.00012) | 31 |
| D1168D | 1 (0.00012) | 21 |
| T1173T | 1 (0.00012) | 36 |
| T1211T | 1 (0.00012) | 71 |
| I1222I | 1 (0.00012) | 38 |
| L1226L | 1 (0.00012) | 26 |
| Y1265Y | 1 (0.00012) | 68 |
| V1279V | 1 (0.00012) | 26 |
| S1299S | 2 (0.00023) | 35 |
| L1328L | 2 (0.00023) | 52 |
| C1337C | 2 (0.00023) | 25 |
| S1372S | 729 (0.085) | 49 |
| P1373P | 1 (0.00012) | 46 |
| L1402L | 1 (0.00012) | 42 |
| L1404L | 1 (0.00012) | 48 |
| A1420A | 1 (0.00012) | 96 |
| L1499L | 1 (0.00012) | 53 |
| R1513R | 1 (0.00012) | 9 |
| I1530I | 1 (0.00012) | 43 |
| P1547P | 1 (0.00012) | 78 |
| S1638S | 1 (0.00012) | 5 |
| V1648V | 27 (0.0031) | 35 |
| G1674G | 1 (0.00012) | 39 |
| P1697P | 1 (0.00012) | 25 |
| Total synonymous variant frequency | 3652 (0.42)* | |
| Collapsed rare (MAF ≤0.01) synonymous variant frequency | 182 (0.042)† |
Number (carrier allele frequency, all synonymous variants).
Number (carrier rare synonymous variant frequency, assuming 1 rare variant per individual).
Table IX.
Synonymous variants identified among infants of African descent in the ESP
| Variant | ESP (n = 2198)* | Distance from splice site, bp |
|---|---|---|
| L9L | 1 (0.00023) | 27 |
| F71F | 1 (0.00023) | 107 |
| A131A | 4 (0.00091) | 55 |
| V150V | 498 (0.11) | 3 |
| A227A | 11 (0.0025) | 68 |
| A229A | 1 (0.00023) | 74 |
| L256L | 1 (0.00023) | 106 |
| T274T | 10 (0.0023) | 52 |
| V282V | 1 (0.00023) | 28 |
| V283V | 1 (0.00023) | 25 |
| L311L | 1 (0.00023) | 58 |
| I318I | 20 (0.0045) | 37 |
| A319A | 1 (0.00023) | 34 |
| P343P | 6 (0.0014) | 39 |
| F353F | 383 (0.087) | 53 |
| F449F | 1 (0.00023) | 62 |
| T574T | 14 (0.0032) | 20 |
| P585P | 329 (0.075) | 14 |
| V623V | 59 (0.013) | 28 |
| C744C | 2 (0.00045) | 32 |
| S747S | 2 (0.00045) | 23 |
| A756A | 3 (0.00068) | 5 |
| P770P | 1 (0.00023) | 47 |
| H780H | 19 (0.0043) | 75 |
| A792A | 3 (0.00068) | 39 |
| S866S | 3 (0.00068) | 85 |
| L897L | 1 (0.00023) | 10 |
| L931L | 1 (0.00023) | 93 |
| V932V | 16 (0.0036) | 96 |
| T966T | 1 (0.00023) | 107 |
| L1001L | 2 (0.00045) | 2 |
| V1037V | 1 (0.00023) | 107 |
| L1065L | 8 (0.0018) | 84 |
| E1093E | 3 (0.00068) | 1 |
| A1119A | 1 (0.00023) | 79 |
| H1137H | 2 (0.00045) | 73 |
| T1211T | 1 (0.00023) | 71 |
| H1247H | 1 (0.00023) | 38 |
| A1280A | 1 (0.00023) | 23 |
| S1299S | 1 (0.00023) | 35 |
| A1311A | 2 (0.00045) | 71 |
| C1315C | 2 (0.00045) | 83 |
| S1372S | 968 (0.22) | 49 |
| P1373P | 2 (0.00045) | 46 |
| A1398A | 2 (0.00045) | 30 |
| A1405A | 1 (0.00023) | 51 |
| S1516S | 1 (0.00023) | 1 |
| D1539D | 1 (0.00023) | 70 |
| A1623A | 1 (0.00023) | 41 |
| V1648V | 5 (0.0011) | 35 |
| S1684S | 2 (0.00045) | 64 |
| T1699T | 1 (0.00023) | 19 |
| Total synonymous variant frequency | 2404 (0.55)* | |
| Collapsed rare (MAF ≤0.01) synonymous variant frequency | 167 (0.076)† |
Number (carrier allele frequency, all synonymous variants).
Number (carrier rare synonymous variant frequency, assuming 1 rare variant per individual).
Discussion
ABCA3 mediates the transport of phospholipids into lamellar bodies and is required for their formation.39–41 Functional studies in surrogate cell systems have suggested that ABCA3 mutations alter intracellular trafficking, impair adenosine triphosphate hydrolysis,22,23,26 or induce endoplasmic reticulum stress.42 Functional studies in other genes have suggested that synonymous variants can affect splicing, alter messenger RNA stability, affect translation rates by altering codon usage, and change protein folding patterns.14
Whereas previous studies identified associations with p.F353F and p.P585P, 2 of the most common synonymous variants in ABCA3, we did not find a statistically significant association for either of these variants with the risk of neonatal RDS.19,20 Possible explanations for our failure to replicate those previously reported findings include population differences (gestational age and ethnicity) between the cohorts. For example, to reduce confounding of developmental disruption of surfactant function, we studied more mature infants (≥34 weeks gestation), whereas the previous studies evaluated preterm infants (25–34 weeks gestation).19,20 A gene-by-development interaction specific for more immature infants might account for their results. Our study focused on infants of European and African descent, whereas the previous studies involved infants of Chinese and Finnish descent. Finally, their large cohort size (n = 503) decreased the risk of type II error, which can confound studies of disease associations of common variants (MAF ≥0.01) that are not under selection pressure and remain in the population.
Our finding that no common (MAF ≥0.01) synonymous variants deviated from Hardy-Weinberg equilibrium suggests stable genetic variation and lack of selection. The similarity between synonymous ABCA3 variant frequencies in our infants without RDS and individuals in the ESP suggests no loss of alleles between the neonatal period and adulthood. Our failure to find a statistically significant interaction between common synonymous variants and rare deleterious ABCA3 mutations suggests the absence of modifying effects on ABCA3 function; however, in vitro studies would permit more reliable, functional testing of variant–mutation interaction.
Alterations in splicing caused by synonymous variants are well described in numerous disease states.43 We used the in silico splice site predictors embedded within Alamut to determine whether synonymous variants disrupted intron–exon splicing patterns. Likely related to the distance of most synonymous variants (>30 base pairs) from the intron–exon junction, no synonymous variants in ABCA3 were predicted to affect splicing with high certainty.44 This lack of splicing effect is consistent with the observation that synonymous variants do not increase the risk of RDS; however, because these predictions are based on computer modeling and not on studies in model systems, we may be misestimating their effects on splicing.
In addition to synonymous and nonsynonymous variants, other genetic alterations can affect gene expression. For example, variants within the promoter, introns, and 5′ and 3′ untranslated regions,45 as well as neighboring deletions46 involving key regulatory elements, can alter gene transcription and translation.
In conclusion, we found no statistical associations between individual common or collapsed rare synonymous ABCA3 variants and the risk of neonatal RDS. We have outlined a computational strategy for assessment of individual synonymous variants discovered during the evaluation of infants with progressive respiratory failure. Future experiments in human model systems may be helpful in evaluating the contributions of individual synonymous variants to disruption of protein function.
Acknowledgments
Supported by the National Institutes of Health (R01 HL065174 [to F.C. and A.H.], R01 HL082747 [to F.C. and A.H.), K12 HL089968 [to F.C.], K08 HL105891 [to J.W.], K08 CA140720-01A1 [to T.D.]), the Children’s Discovery Institute (to R.M. and T.D.), the Saigh Foundation (to F.C. and A.H.), the Keck Foundation (to R.M.), and Kailos Genetics (to R.M.). The National Heart, Lung, and Blood Institute GO Exome Sequencing Project and its ongoing studies produced and provided exome variant calls for comparison: the Lung GO Sequencing Project (HL-102923), the WHI Sequencing Project (HL-102924), the Broad GO Sequencing Project (HL-102925), the Seattle GO Sequencing Project (HL-102926), and the Heart GO Sequencing Project (HL-103010).
The authors would like to acknowledge Leslie Walther, RN, and Rosina Schiff, BS, for their significant contributions to patient recruitment.
Glossary
- ABCA3
Adenosine triphosphate-binding cassette A3 transporter
- ESP
Exome Sequencing Project
- MAF
Minor allele frequency
- RDS
Respiratory distress syndrome
Footnotes
The authors declare no conflicts of interest.
References
- 1.Avery ME, Mead J. Surface properties in relation to atelectasis and hyaline membrane disease. Am J Dis Child. 1959;97:517–523. doi: 10.1001/archpedi.1959.02070010519001. [DOI] [PubMed] [Google Scholar]
- 2.Levit O, Jiang Y, Bizzarro MJ, Hussain N, Buhimschi CS, Gruen JR, et al. The genetic susceptibility to respiratory distress syndrome. Pediatr Res. 2009;66:693–697. doi: 10.1203/PDR.0b013e3181bbce86. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Nogee L, de Mello DE, Dehner LP, Colten HR. Brief report: deficiency of pulmonary surfactant protein B in congenital alveolar proteinosis. N Engl J Med. 1993;328:406–410. doi: 10.1056/NEJM199302113280606. [DOI] [PubMed] [Google Scholar]
- 4.Nogee LM, Dunbar AE, III, Wert SE, Askin F, Hamvas A, Whitsett JA. A mutation in the surfactant protein C gene associated with familial interstitial lung disease. N Engl J Med. 2001;344:573–579. doi: 10.1056/NEJM200102223440805. [DOI] [PubMed] [Google Scholar]
- 5.Shulenin S, Nogee LM, Annilo T, Wert SE, Whitsett JA, Dean M. ABCA3 gene mutations in newborns with fatal surfactant deficiency. N Engl J Med. 2004;350:1296–1303. doi: 10.1056/NEJMoa032178. [DOI] [PubMed] [Google Scholar]
- 6.Wambach JA, Yang P, Wegner DJ, An P, Hackett BP, Cole FS, et al. Surfactant protein-C promoter variants associated with neonatal respiratory distress syndrome reduce transcription. Pediatr Res. 2010;68:216–220. doi: 10.1203/PDR.0b013e3181eb5d68. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Wambach JA, Wegner DJ, Depass K, Heins H, Druley TE, Mitra RD, et al. Single ABCA3 mutations increase risk for neonatal respiratory distress syndrome. Pediatrics. 2012;130:e1575–e1582. doi: 10.1542/peds.2012-0918. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Dean M, Hamon Y, Chimini G. The human ATP-binding cassette (ABC) transporter superfamily. J Lipid Res. 2001;42:1007–1017. [PubMed] [Google Scholar]
- 9.Yamano G, Funahashi H, Kawanami O, Zhao LX, Ban N, Uchida Y, et al. ABCA3 is a lamellar body membrane protein in human lung alveolar type II cells. FEBS Lett. 2001;508:221–225. doi: 10.1016/s0014-5793(01)03056-3. [DOI] [PubMed] [Google Scholar]
- 10.Mulugeta S, Gray JM, Notarfrancesco KL, Gonzales LW, Koval M, Feinstein SI, et al. Identification of LBM180, a lamellar body limiting membrane protein of alveolar type II cells, as the ABC transporter protein ABCA3. J Biol Chem. 2002;277:22147–22155. doi: 10.1074/jbc.M201812200. [DOI] [PubMed] [Google Scholar]
- 11.Bullard J, Wert SE, Nogee LM. ABCA3 deficiency: neonatal respiratory failure and interstitial lung disease. Semin Perinatol. 2006;30:327–334. doi: 10.1053/j.semperi.2005.12.001. [DOI] [PubMed] [Google Scholar]
- 12.Brasch F, Schimanski S, Muhlfeld C, Barlage S, Langmann T, Aslanidis C, et al. Alteration of the pulmonary surfactant system in full-term infants with hereditary ABCA3 deficiency. Am J Respir Crit Care Med. 2006;174:571–580. doi: 10.1164/rccm.200509-1535OC. [DOI] [PubMed] [Google Scholar]
- 13.Chamary JV, Parmley JL, Hurst LD. Hearing silence: non-neutral evolution at synonymous sites in mammals. Nat Rev Genet. 2006;7:98–108. doi: 10.1038/nrg1770. [DOI] [PubMed] [Google Scholar]
- 14.Sauna ZE, Kimchi-Sarfaty C. Understanding the contribution of synonymous mutations to human disease. Nat Rev Genet. 2011;12:683–691. doi: 10.1038/nrg3051. [DOI] [PubMed] [Google Scholar]
- 15.Plotkin JB, Kudla G. Synonymous but not the same: the causes and consequences of codon bias. Nat Rev Genet. 2011;12:32–42. doi: 10.1038/nrg2899. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Cartegni L, Chew SL, Krainer AR. Listening to silence and understanding nonsense: exonic mutations that affect splicing. Nat Rev Genet. 2002;3:285–298. doi: 10.1038/nrg775. [DOI] [PubMed] [Google Scholar]
- 17.Kimchi-Sarfaty C, Oh JM, Kim IW, Sauna ZE, Calcagno AM, Ambudkar SV, et al. A “silent” polymorphism in the MDR1 gene changes substrate specificity. Science. 2007;315:525–528. doi: 10.1126/science.1135308. [DOI] [PubMed] [Google Scholar]
- 18.Tsai CJ, Sauna ZE, Kimchi-Sarfaty C, Ambudkar SV, Gottesman MM, Nussinov R. Synonymous mutations and ribosome stalling can lead to altered folding pathways and distinct minima. J Mol Biol. 2008;383:281–291. doi: 10.1016/j.jmb.2008.08.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Karjalainen MK, Haataja R, Hallman M. Haplotype analysis of ABCA3: association with respiratory distress in very premature infants. Ann Med. 2008;40:56–65. doi: 10.1080/07853890701611094. [DOI] [PubMed] [Google Scholar]
- 20.Jiang L, Wu YD, Xu XF, Du LZ. Polymorphism analysis of the ABCA3 gene: association with neonatal respiratory distress syndrome in preterm infants. Chin Med J. 2012;125:1594–1598. [PubMed] [Google Scholar]
- 21.Hamvas A, Faro A. Lung transplantation for inherited disoders of surfactant metabolism. NeoReviews. 2008;9:e468–e476. [Google Scholar]
- 22.Matsumura Y, Ban N, Inagaki N. Aberrant catalytic cycle and impaired lipid transport into intracellular vesicles in ABCA3 mutants associated with nonfatal pediatric interstitial lung disease. Am J Physiol Lung Cell Mol Physiol. 2008;295:L698–L707. doi: 10.1152/ajplung.90352.2008. [DOI] [PubMed] [Google Scholar]
- 23.Matsumura Y, Ban N, Ueda K, Inagaki N. Characterization and classification of ATP-binding cassette transporter ABCA3 mutants in fatal surfactant deficiency. J Biol Chem. 2006;281:34503–34514. doi: 10.1074/jbc.M600071200. [DOI] [PubMed] [Google Scholar]
- 24.Ng PC, Henikoff S. SIFT: predicting amino acid changes that affect protein function. Nucleic Acids Res. 2003;31:3812–3814. doi: 10.1093/nar/gkg509. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Sunyaev S, Ramensky V, Koch I, Lathe W, III, Kondrashov AS, Bork P. Prediction of deleterious human alleles. Hum Mol Genet. 2001;10:591–597. doi: 10.1093/hmg/10.6.591. [DOI] [PubMed] [Google Scholar]
- 26.Cheong N, Madesh M, Gonzales LW, Zhao M, Yu K, Ballard PL, et al. Functional and trafficking defects in ATP binding cassette A3 mutants associated with respiratory distress syndrome. J Biol Chem. 2006;281:9791–9800. doi: 10.1074/jbc.M507515200. [DOI] [PubMed] [Google Scholar]
- 27.Druley TE, Vallania FL, Wegner DJ, Varley KE, Knowles OL, Bonds JA, et al. Quantification of rare allelic variants from pooled genomic DNA. Nat Methods. 2009;6:263–265. doi: 10.1038/nmeth.1307. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Vallania FL, Druley TE, Ramos E, Wang J, Borecki I, Province M, et al. High-throughput discovery of rare insertions and deletions in large cohorts. Genome Res. 2010;20:1711–1718. doi: 10.1101/gr.109157.110. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Desmet FO, Hamroun D, Lalande M, Collod-Beroud G, Claustres M, Beroud C. Human Splicing Finder: an online bioinformatics tool to predict splicing signals. Nucleic Acids Res. 2009;37:e67. doi: 10.1093/nar/gkp215. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Pertea M, Lin X, Salzberg SL. GeneSplicer: a new computational method for splice site prediction. Nucleic Acids Res. 2001;29:1185–1190. doi: 10.1093/nar/29.5.1185. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Yeo G, Burge CB. Maximum entropy modeling of short sequence motifs with applications to RNA splicing signals. J Comput Biol. 2004;11:377–394. doi: 10.1089/1066527041410418. [DOI] [PubMed] [Google Scholar]
- 32.Reese MG, Eeckman FH, Kulp D, Haussler D. Improved splice site detection in Genie. J Comput Biol. 1997;4:311–323. doi: 10.1089/cmb.1997.4.311. [DOI] [PubMed] [Google Scholar]
- 33.Hellen B. Splice site tools: a comparative analysis report. Mancester: National Genetics Reference Laboratory. 2009:1–12. [Google Scholar]
- 34.Li B, Leal SM. Methods for detecting associations with rare variants for common diseases: application to analysis of sequence data. Am J Hum Genet. 2008;83:311–321. doi: 10.1016/j.ajhg.2008.06.024. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Kelly M, Semsarian C. Multiple mutations in genetic cardiovascular disease: a marker of disease severity? Circ Cardiovasc Genet. 2009;2:182–190. doi: 10.1161/CIRCGENETICS.108.836478. [DOI] [PubMed] [Google Scholar]
- 36.Anadkat JS, Kuzniewicz MW, Chaudhari BP, Cole FS, Hamvas A. Increased risk for respiratory distress among white, male, late preterm and term infants. J Perinatol. 2012;32:780–785. doi: 10.1038/jp.2011.191. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Hook B, Kiwi R, Amini SB, Fanaroff A, Hack M. Neonatal morbidity after elective repeat cesarean section and trial of labor. Pediatrics. 1997;100:348–353. doi: 10.1542/peds.100.3.348. [DOI] [PubMed] [Google Scholar]
- 38.Kruglyak L, Nickerson DA. Variation is the spice of life. Nat Genet. 2001;27:234–236. doi: 10.1038/85776. [DOI] [PubMed] [Google Scholar]
- 39.Ban N, Matsumura Y, Sakai H, Takanezawa Y, Sasaki M, Arai H, et al. ABCA3 as a lipid transporter in pulmonary surfactant biogenesis. J Biol Chem. 2007;282:9628–9634. doi: 10.1074/jbc.M611767200. [DOI] [PubMed] [Google Scholar]
- 40.Nagata K, Yamamoto A, Ban N, Tanaka AR, Matsuo M, Kioka N, et al. Human ABCA3, a product of a responsible gene for abca3 for fatal surfactant deficiency in newborns, exhibits unique ATP hydrolysis activity and generates intracellular multilamellar vesicles. Biochem Biophys Res Commun. 2004;324:262–268. doi: 10.1016/j.bbrc.2004.09.043. [DOI] [PubMed] [Google Scholar]
- 41.Matsumura Y, Sakai H, Sasaki M, Ban N, Inagaki N. ABCA3-mediated choline-phospholipids uptake into intracellular vesicles in A549 cells. FEBS Lett. 2007;581:3139–3144. doi: 10.1016/j.febslet.2007.05.078. [DOI] [PubMed] [Google Scholar]
- 42.Weichert N, Kaltenborn E, Hector A, Woischnik M, Schams A, Holzinger A, et al. Some ABCA3 mutations elevate ER stress and initiate apoptosis of lung epithelial cells. Respir Res. 2011;12:4. doi: 10.1186/1465-9921-12-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Sauna ZE, Kimchi-Sarfaty C, Ambudkar SV, Gottesman MM. Silent polymorphisms speak: how they affect pharmacogenomics and the treatment of cancer. Cancer Res. 2007;67:9609–9612. doi: 10.1158/0008-5472.CAN-07-2377. [DOI] [PubMed] [Google Scholar]
- 44.Chamary JV, Hurst LD. Biased codon usage near intron–exon junctions: selection on splicing enhancers, splice-site recognition or something else? Trends Genet. 2005;21:256–259. doi: 10.1016/j.tig.2005.03.001. [DOI] [PubMed] [Google Scholar]
- 45.Chatterjee S, Pal JK. Role of 5’- and 3’-untranslated regions of mRNAs in human diseases. Biol Cell. 2009;101:251–262. doi: 10.1042/BC20080104. [DOI] [PubMed] [Google Scholar]
- 46.Henderson LB, Melton K, Wert S, Couriel J, Bush A, Ashworth M, et al. Large ABCA3 and SFTPC deletions resulting in lung disease. Ann Am Thorac Soc. 2013;10:602–607. doi: 10.1513/AnnalsATS.201306-170OC. [DOI] [PMC free article] [PubMed] [Google Scholar]