Prenatal antibiotics exposure does not influence experimental allergic asthma in mice

Abstract

Changes in microbiome (dysbiosis) contribute to severity of allergic asthma. Preexisting epidemiological studies in humans correlate perinatal dysbiosis with increased long-term asthma severity. However, these studies cannot discriminate between prenatal and postnatal effects of dysbiosis and suffer from a high variability of dysbiotic causes ranging from antibiotic treatment, delivery by caesarian section to early-life breastfeeding practices. Given that maternal antibiotic exposure in mice increases the risk of newborn bacterial pneumonia in offspring, we hypothesized that prenatal maternal antibiotic-induced dysbiosis induces long-term immunological effects in the offspring that also increase long-term asthma severity. Therefore, dams were exposed to antibiotics (gentamycin, ampicillin, vancomycin) from embryonic day 15 until birth. Six weeks later, asthma was induced in the offspring by repeated applications of house dust mite extract. Airway function, cytokine production, pulmonary cell composition and distribution were assessed. Our study revealed that prenatally induced dysbiosis in mice led to an increase in pulmonary Th17⁺ non-conventional T cells with limited functional effect on airway resistance, pro-asthmatic Th2/Th17 cytokine production, pulmonary localization and cell-cell contacts. These data indicate that dysbiosis-related immune-modulation with long-term effects on asthma development occurs to a lesser extent prenatally and will allow to focus future studies on more decisive postnatal timeframes.

Keywords: ILCs; RORγt; Th17 response; asthma; dysbiosis; microbiome.

Figure 1

Prenatal antibiotic exposure increases the frequency of IL-17A⁺ pulmonary cells without affecting gene expression. (A) C57BL/6 mice were treated with vancomycin, ampicillin and gentamycin (all at 1.0 mg/mL)(Abx) from prenatal day 15 until birth. Mice treated with sucralose served as controls (ctrl.). At 6-8 weeks, offspring C57BL/6 mice were immunized i.p. with 10 µg/100 µL of HDM. Seven days later mice received an additional i.p. injection of HDM. After 14 and 21 days mice were treated with 100 µg/40 µL HDM i.t. Mice treated with PBS served as non-asthmatic controls (B, D). 72 h after the last treatment the asthma phenotype was assessed. (B) Single lung cell suspensions were restimulated with PMA (50 ng/mL) and ionomycin (0.5 mg/mL). After intracellular staining of IL-17A, CD90.2⁺ lymphocytes were analyzed regarding IL-17A production by flow cytometry. (C) To further differentiate IL-17A⁺ populations, CD3 and TCRβ expression was analyzed to identify CD3⁺TCRβ⁺ αβ T cells, CD3⁺TCRβ^- non-conventional T cells and CD3^-TCRβ^- ILCs. (D) Further, 72 h after the last in vivo treatment RNA expression of Th17-related cytokines normalized to the expression of the housekeeping gene S14 by pulmonary cells was assessed. Scatter plot with mean ± SEM; n = 5-25 mice. Each dot represents one mouse. Differences between groups were tested by one-way ANOVA (after passing D’Agostino and Pearson omnibus normality Test) and Tukey Post Test)(B), Student T Test (after passing D’Agostino and Pearson omnibus normality Test)(C)or Kruskal-Wallis-Test and Dunns’s Post Test for significance (D); * p < 0.05, ** p < 0.01.

Figure 2

Prenatal antibiotic exposure of dams does not influence the asthma phenotype of the offspring. C57BL/6 mice were treated with vancomycin, ampicillin and gentamycin (all at 1.0 mg/mL)(Abx) from prenatal day 15 until birth. Mice treated with sucralose served as controls (ctrl.). At 6-8 weeks, offspring C57BL/6 mice were immunized i.p. with 10 µg/100 µL of HDM. Seven days later mice received an additional i.p. injection of HDM. After 14 and 21 days, mice were treated with 100 µg/40 µL HDM i.t. Mice treated with PBS served as non-asthmatic controls. 72 h after the last treatment AHR in response to i.t. administration of methacholine measured as airway resistance using Flexivent (A), cellular infiltration of the airways (B, C) and cytokine production of single lung cell suspensions after 72 h ex vivo cell culture (D) were assessed. Scatter plot with mean ± SEM, n = 6-18 mice. Each dot represents one mouse. Dotted lines (D) indicate the detection limit. Differences between groups were tested by two-way ANOVA and Bonferroni Post Test (A) or Kruskal-Wallis-Test and Dunns’s Post Test (B–D) for significance; * p < 0.05, ** p < 0.01, *** p < 0.001.

Figure 3

RORγt is suitable as a marker for pulmonary IL-17A⁺ cells in situ. At 6-8 weeks C57BL/6 mice were immunized i.p. with 10 µg/100 µL of HDM. Seven days later mice received an additional i.p. injection of HDM. After 14 and 21 days mice were treated with 100 µg/40 µL HDM i.t. 72 h after the last treatment 300 µm lung sections of agarose filled lungs were prepared and stained with anti-RORγt (red) and anti-CD3 (green) (A) or anti-RORγt (red) and anti-GATA3 (turquoise) (B). (C) RORγt⁺ cells were analyzed regarding CD3 expression. (D) Total cells were analyzed regarding RORγt and GATA3 expression. Stained slices were evaluated with confocal microscopy. Abbreviations: AW, airway. Data are representative of two independent experiments. Scatter plot with mean ± SEM; n = 7 mice. Each dot represents the Z-stack of one mouse.

Figure 4

Prenatal antibiotic exposure of dams does not influence localization of offspring pulmonary RORγt⁺ cells in situ. C57BL/6 mice were treated with vancomycin, ampicillin and gentamycin (all at 1.0 mg/mL)(Abx) from prenatal day 15 until birth. Mice treated with sucralose served as controls (ctrl.). At 6-8 weeks, offspring C57BL/6 mice were immunized i.p. with 10 µg/100 µL of HDM. Seven days later mice received an additional i.p. injection of HDM. After 14 and 21 days mice were treated with 100 µg/40 µL HDM i.t. 72 h after the last treatment 300 µm lung sections of agarose filled lungs were prepared. (A, B) For visualization of airways and blood vessels anti-aSMA (green) was added. (C) Alveoli were identified by autofluorescence. (D) Stained slices were evaluated with laser scanning confocal microscopy generating Z-Stacks for localization of RORγt⁺ cells (red). Abbreviations: AW, airway; BV, blood vessel. Data are representative of at least four independent experiments. Scatter plot with mean ± SEM. n = 4-5 mice. Each dot represents the Z-stack of one mouse.

Figure 5

Prenatal antibiotic exposure of dams does not influence localization of offspring pulmonary RORγt⁺ subpopulations in situ. C57BL/6 mice were treated with vancomycin, ampicillin and gentamycin (all at 1.0 mg/mL)(Abx) from prenatal day 15 until birth. Mice treated with sucralose served as controls (ctrl.). At 6-8 weeks, offspring C57BL/6 mice were immunized i.p. with 10 µg/100 µL of HDM. Seven days later mice received an additional i.p. injection of HDM. After 14 and 21 days mice were treated with 100 µg/40 µL HDM i.t. 72 h after the last treatment 300 µm lung sections of agarose filled lungs were prepared. Staining with anti-RORγt (red), anti-CD3 (turquoise) and anti-TCRβ (green) allowed discrimination of CD3⁺TCRβ⁺ αβ T cells (A), CD3⁺TCRβ^- non-conventional T cells (B) and CD3^-TCRβ^- ILC3s (C). Stained slices were evaluated with laser scanning confocal microscopy generating Z-Stacks. (D) A representative overview of a large airway is shown. Frequencies of αβ T cells, non-conventional T cells (nc T) and ILC3s among RORγt⁺ cells were determined in total lungs (E) around large (F) and small airways/blood vessels (G). Abbreviations: AW, airway. Data are representative of at least four independent experiments. Scatter plot with mean ± SEM. n = 4-5 mice. Each dot represents the Z-stack of one mouse.

Figure 6

Prenatal antibiotic exposure of dams does not influence the frequency of contacts between RORγt⁺ cells and CD11c⁺ cells in situ. C57BL/6 mice were treated with vancomycin, ampicillin and gentamycin (all at 1.0 mg/mL)(Abx) from prenatal day 15 until birth. Mice treated with sucralose served as controls (ctrl.). At 6-8 weeks, offspring C57BL/6 mice were immunized i.p. with 10 µg/100 µL of HDM. Seven days later mice received an additional i.p. injection of HDM. After 14 and 21 days mice were treated with 100 µg/40 µL HDM i.t. 72 h after the last treatment 300 µm lung sections of agarose filled lungs were prepared. Staining with anti-RORγt (red) and anti-CD11c (turquoise) allowed evaluation of interactions between RORγt⁺ cells and CD11c⁺ DCs. Stained slices were evaluated with laser scanning confocal microscopy generating Z-Stacks. (A) A representative overview of a large airway is shown. Enlarged 3D views (B) of framed areas allowed 3D analysis of contact rates of RORγt⁺ cells with CD11c⁺ DCs (C). Abbreviations: AW, airway. Data are representative of at least four independent experiments. Scatter plot with mean ± SEM. n = 4 mice. Each dot represents the Z-stack of one mouse.