Maternal and infant microbiome: next-generation indicators and targets for intergenerational health and nutrition care

Abstract

Microbes are commonly sensitive to shifts in the physiological and pathological state of their hosts, including mothers and babies. From this perspective, the microbiome may be a good indicator for diseases during pregnancy and has the potential to be used for perinatal health monitoring. This is embodied in the application of microbiome from multi body sites for auxiliary diagnosis, early prediction, prolonged monitoring, and retrospective diagnosis of pregnancy and infant complications, as well as nutrition management and health products developments of mothers and babies. Here we summarized the progress in these areas and explained that the microbiome of different body sites is sensitive to different diseases and their microbial biomarkers may overlap between each other, thus we need to make a diagnosis prudently for those diseases. Based on the microbiome variances and additional anthropometric and physical data, individualized responses of mothers and neonates to meals and probiotics/prebiotics were predictable, which is of importance for precise nutrition and probiotics/prebiotics managements and developments. Although a great deal of encouraging performance was manifested in previous studies, the efficacy could be further improved by combining multi-aspect data such as multi-omics and time series analysis in the future. This review reconceptualizes maternal and infant health from a microbiome perspective, and the knowledge in it may inspire the development of new options for the prevention and treatment of adverse pregnancy outcomes and bring a leap forward in perinatal health care.

Keywords: disease detection; health care products; microbiome; newborn; pregnancy management.

Figure 1.

Overview of maternal and neonatal microbiome utility in perinatal health care. (A) Microbiome in oral, stool, and vagina are instructive for multiple pregnancy-complications diagnose, prediction, and monitoring. Combining of microbiome in multi body site could efficiently contribute to the early detection of pregnant diseases. (B) Prevention of fetal overgrowth and control of maternal hyperglycemia are the primary goal for GDM treatment, which is usually achieved by dietary modification and promotion of physical activity to minimize postprandial glucose elevations. Encouragingly, machine-learning model trained based on maternal microbiome, meal composition, and genetics data could accurately predict postprandial glycemic responses, which may lay the foundation for precise pregnancy management. (C) The infant microbiome can be used as an indicator and target of flora maturity and nutritional health status, supporting normal immune development and reducing the risk of immune-mediated disease. (D) Multi-omics including microbiome contributes to the prebiotics and probiotics discovering and safety evaluation. The icons in the central loop represent the main data types that used for the relevant sections of maternal and neonatal health care. The markers under each icon represent the application scenario of this type of data. For pregnancy complications diagnose, the datasets including microbiome from oral, stool, and vagina; for pregnancy precise management, the datasets including stool microbiome, physical activity, host genetics, and diets compositions; for neonatal health care, the datasets including stool microbiome, body length, and clinical records data; for health products development, the datasets including stool microbiome and host genetics.

Figure 2.

Features employed in the machine-learning model for the prediction of postprandial glycemic responses. The features are mainly classified into six groups: (A) baseline characteristics, (B) genetics, (C) gut microbiome features, (D) questionnaires, (E) meal context, and (F) meal composition. Meal context represents the activity or snacks consumption that may influence the glycemic responses after meal. BMI: body mass index; HR: heart rate; Fat/carbs: ratio of fat to carbohydrates content in meals composition; GWAS: genome wide association study.

Figure 3.

Measurement of developmental status and effect assessment of infants with microbiome. (A) The newborns in extreme undeveloped country or town tends to have immature gut microbiome characterized by absent B. infantis in gut, due to lower HMOs intake. Depleted bifidobacteria is associated with systemic inflammation and immune imbalance early in life. Immature gut microbiome may affect the growth, long-term health, and the susceptibility to allergy. (B) Bifidobacterium infantis metabolizes HMOs to indole-3-lactic acid (ILA), which skew naive T cells away from proinflammatory Th17 and Th2 and toward anti-inflammatory Th1. IL-27 limits Th2- and Th17-type responses and regulate T cell function by activating Th1. Galectin-1 induces L-27 and IL-10 and act through IFNβ-dependent reprogramming of tissue macrophages and be essential for inflammation resolving.

Figure 4.

Differences of the prebiotic selectivity with the methods based on in vitro culture and microbiome. (A) For traditional method, the prebiotics authentication only based on that whether this substrate could promote the proliferation of probiotics such as Bifidobacterium and Lactobacillus. (B) However, using microbiome, we could comprehensively assess the probiotic effect in real gut microbiome community and screen out the genuine prebiotics.