Markers in Infants of Mothers With Asthma and Associations With Respiratory Outcomes

Abstract

Background: In utero mechanisms related to oxidative stress response, inflammation, and extracellular matrix turnover may influence fetal lung development in the offspring of asthmatic mothers. Therefore, we aimed to determine whether levels of cord blood proteins differ between the offspring of asthmatic and nonasthmatic mothers. In addition, we aimed to examine if these proteins are associated with infant lung function, bronchiolitis hospitalization in infancy, and asthma at six years.

Methods: We compared protein levels of infants (n = 715) from the Swiss Basel-Bern Infant Lung Development and the Australian Breathing for Life Trial birth cohorts using Tobit regression and network analyses. Using adjusted linear and logistic regression, we determined their association with postnatal infant lung function, bronchiolitis hospitalization in infancy, and asthma at six years.

Results: Infants born to asthmatic mothers had lower levels of matrix metalloproteinase-9 (MMP-9, β-coefficient [β] -0.67, 95% confidence interval [-1.07; -0.27] p_adj = 0.009) and Interferon gamma (IFN-γ, β -0.77 [-1.21; -0.32], p_adj = 0.009), and higher levels of p62 (β 1.15 [0.30; 2.00], p_adj = 0.030). p62 levels were inversely associated with minute ventilation (β -16.18 [-28.44; -3.91], p_adj = 0.032). Functional residual capacity values were inversely associated with both IFN-γ (β -1.26, [-2.41; -0.11], p_adj = 0.063) and MMP-9 levels (β -1.27, [-2.53; -0.01], p_adj = 0.063). MMP-9 levels were inversely associated with both the risk of bronchiolitis hospitalization (odds ratio 0.47, [0.29; 0.77], p_adj = 0.004) and the risk of asthma (aOR 0.53, 95% CI, 0.32-0.86, p_adj = 0.033).

Conclusions: Protein levels differed between offspring of asthmatic and non-asthmatic mothers. These markers were associated with postnatal lung function, bronchiolitis hospitalization, and asthma.

Keywords: asthma; asthma during pregnancy; bronchiolitis; infant lung function; proteins.

FIGURE 1

Cord blood protein levels in infants born to a mother with asthma during pregnancy. Beclin‐1, Beclin‐1; ECM‐turnover, extracellular matrix turnover; IFN‐γ, Interferon gamma; MMP‐9, matrix metalloproteinase 9; p62, ubiquitin‐binding protein sequestosome 1. Estimates are reported as a β‐coefficient and represent results from a Tobit regression model, where maternal asthma during pregnancy is the primary exposure. Estimates are reported as coefficients with 95% CI log2‐transformed protein levels, also adjusted for sex, having siblings at birth, gestational age, maternal smoking during pregnancy, mode of delivery, birth weight in z‐score, study center, and time of cord blood processing. p _adj‐value * < 0.05; **p _adj‐value < 0.01.

FIGURE 2

Correlation network in infants born to asthmatic mothers (A) and infants born to non‐asthmatic mothers (B). ATG5, autophagy protein 5; Beclin‐1, Beclin‐1; EGF, Epidermal Growth Factor; IFN‐γ, Interferon gamma; IL‐13, interleukin 13; IL‐17A, interleukin 17A; IL‐1β, interleukin 1β; IL‐4, interleukin 4; IL‐8, interleukin 8; MMP‐9, matrix metalloproteinase 9; p62, ubiquitin‐binding protein sequestosome 1; PDGF‐AA, platelet‐derived growth factor AA; SIRT1, Sirtuin 1; TGF‐β1, transforming growth factor β1; TNFα, tumor necrosis factor α; VEGF‐A, Vascular endothelial growth factor A. Measures of centrality are shown as connections (edges) are established using Spearman's correlation coefficient. Correlations are displayed when the absolute correlation coefficient is > 0.3. Each protein is represented by a node, and edge colors reflect the strength of correlation between proteins.

FIGURE 3

Difference in measures of centrality according to protein markers for infants born to asthmatics and infants born to non‐asthmatics. ATG5, autophagy protein 5; Beclin‐1, Beclin‐1; EGF, epidermal growth factor; IFN‐γ, interferon gamma; IL‐13, interleukin 13; IL‐17A, interleukin 17A; IL‐1β, interleukin 1β; IL‐4, interleukin 4; IL‐8, interleukin 8; MMP‐9, matrix metalloproteinase 9; p62, ubiquitin‐binding protein sequestosome 1; PDGF‐AA, platelet‐derived growth factor AA; SIRT1, Sirtuin 1; TGF‐β1, transforming growth factor β1; TNFα, tumor necrosis factor α; VEGF‐A, vascular endothelial growth factor A. Graph shows network analysis centrality measures including, in red showing the degree centrality (which is the number of connections a node has), in green the strength which refers to the magnitude of connection between nodes, and in blue the betweenness centrality representing the frequency of a node acting as a bridge.

FIGURE 4

Coefficient estimates from multivariable linear regression models showing the associations between three representative markers as exposures and lung function parameters measured at 4–6 weeks. ECM‐turnover, extracellular matrix turnover; FRC, functional residual capacity; IFN‐γ, interferon gamma; LCI, lung clearance index; MMP‐9, matrix metalloproteinase 9; p62, ubiquitin‐binding protein sequestosome 1; tPTEF/tE%, the ratio of time to peak tidal expiratory flow as a percentage of total expiratory time. Estimates are reported as a coefficient with corresponding 95% CI confidence interval according to protein levels as exposure and adjusted for maternal asthma during pregnancy, sex, having siblings at birth, postmenstrual age at lung function test in weeks, weight at test, maternal smoking during pregnancy, mode of delivery, study centre, and time of cord blood processing in days. p‐values shown are after Benjamini–Hochberg multiple comparison correction.