PRENATAL ANTIBIOTIC THERAPY PREDISPOSES TO NEONATAL SEPSIS INDUCED BY E.COLI: A ROLE FOR BREAST MILK IGA?

Antibiotics are administered to 20-30% of pregnant women, to treat maternal infections or to prevent fetal and neonatal bacterial colonization in selected cases of high infection risk, such as premature rupture of membranes (PROM). Despite the administration of antibiotics to neonates, i.e. in the first days of life, is known to increase the risk for subsequent late-onset sepsis (LOS), necrotizing enterocolitis (NEC), and possibly of long term adverse outcomes such as obesity and allergic manifestations, the effect of prenatal antibiotics on neonatal immune development and function is poorly understood. Here, using a murine model of prenatally-restricted antibiotic therapy (Abx), we show that newborn mice born to mothers previously treated with Abx are more susceptible to LOS caused by Escherichia coli, in correlation with an increased bacterial translocation to distant organs (mesenteric lymph nodes, spleen, liver). Increased neonatal mortality after maternal Abx did not correlate with major modifications in blood or spleen populations of immune cells (neutrophils, activated monocytes, CD4+ and CD8+ T cells), nor with changes in gut epithelial structure and integrity, but correlated with a significantly lower amount of fecal IgA, stomach IgA, IgA-coating of intestinal bacteria, and IgA production by plasma cells of the maternal mammary gland. Cross fostering experiments reinforced the evidence that low breast milk IgA after maternal Abx was a key factor to increase the mortality rate of pups. Moreover, the adverse effect of prenatal Abx persisted after the interruption of breastfeeding, in weaned young adults. At 28 days of life, prenatal Abx caused a reduction in IgA+ plasma cells, both CD11b+ and CD11b-, in the lamina propria (LP) of the ileum, a reduction in faecal sIgA, and decreased the amount of colonic T-regs and of Th17-like lymphocytes (CD4+ RORgt+ antigen-specific T cells). Maternal supplementation with postbiotics during the last 5 days of pregnancy was not sufficient to revert the antibiotic-induced phenotype, at least in our experimental setting. Our data highlight the importance of breast milk IgA for the prevention of neonatal LOS, and the influence of prenatal Abx. If confirmed on human samples, these results may lay the foundation for screening strategies of IgA in breast milk, especially in selected high-risk situations such as prenatal Abx administration, and/or extremely preterm neonates receiving maternal (or donor) breast milk.