Maternal microbiota in pregnancy and early life

See allHide authors and affiliations

Science  06 Sep 2019:
Vol. 365, Issue 6457, pp. 984-985
DOI: 10.1126/science.aay0618

A microbial bond is shared between mother and child that renders the age-old dichotomy of “nature versus nurture” obsolete. Molded by both host biology and the physical relationship between mother and child, this microbial connection is formed at the earliest moments of life when the newborn's skin and mucosal surfaces are seeded with microorganisms that inhabit the mother's body, referred to as the maternal microbiota. This initial microbial exposure establishes an early-life microbiota that engages in a mutualistic relationship with the host, and leaves a lasting impression on childhood development that can control the balance between health and disease. The quest to understand this microbial bond has uncovered exciting new discoveries about host–microbial mutualism and immune development in early life, while simultaneously revolutionizing our understanding of how certain traits and diseases are passed down through generations.

In humans and other mammals, the first microbes encountered in early life are those from the maternal microbiota. Despite the foundational nature of this event in human development, uncertainty persists about the precise timing of this first contact. The dogmatic belief that fetal development occurs within a sterile intrauterine environment has been challenged by evidence of bacterial genomic DNA within placental and chorioamniotic tissues and culturable microbes in newborn meconium (the first feces passed after birth), raising the controversial possibility that microbial colonization may begin in utero (1, 2). However, these observations stop short of providing definitive evidence of colonization during fetal development with a bona fide microbiota (that is, a live, persistent, and functional community of microorganisms), and thus the concept of a fetal microbiota remains the subject of debate. In support of this, a recent study found that the human placenta was devoid of a microbiome although it was found to contain potential pathogens in a small proportion of samples (3).

On the contrary, a wealth of evidence demonstrates that the early-life microbiota is seeded at the time of delivery through contact with maternal commensal (natural, nonpathogenic) bacteria that inhabit the birth canal. The microbial inoculum of vaginally delivered neonates is dominated by maternal cervicovaginal and fecal microbes, whereas delivery by cesarian section imparts distinct microbial consortia that are often dominated by skin microorganisms (46). These findings, together with data from observational studies suggesting an association between cesarian delivery and increased risk of adverse childhood health outcomes (obesity, asthma, and others), have led some to hypothesize that the distinctive early-life microbiota associated with cesarian delivery may have lasting effects on childhood health (7). Although this does not constitute a causal relationship, this provocative association has contributed to the increasingly popular practice of “vaginal seeding” of infants born by cesarian section (exposing neonates to maternal vaginal content in the first minutes of life) in an attempt to recapitulate the microbial exposure of a vaginal delivery (8). However, observational studies of maternal–infant pairs have challenged the lasting impact of delivery route on the early-life microbiota with data showing evolution of microbial communities over the first 6 weeks of life that culminates in a diversified microbiota regardless of the original route of delivery (9).

What is clear is that seeding of early-life microbiota with maternal microbes leaves a lasting imprint on the biology of offspring. Many studies, of both humans and mice, have found links between the composition of the maternal microbiota and subsequent risk of childhood illnesses, including infection, asthma, allergy, autoimmunity, and metabolic diseases such as obesity and diabetes (7). Furthermore, the discovery that pathological alterations of microbiota composition and diversity (called “dysbiosis”) can confer susceptibility to a variety of chronic health conditions in adulthood has led investigators to question whether these dysbiosis-associated phenotypes can be vertically transmitted from mother to child via the microbiota. For example, a study of 935 mother–infant pairs from Canada found that children of overweight or obese mothers had a threefold increased risk of becoming obese at 1 year, which increased to fivefold with cesarian delivery (6). The authors identified taxonomic signatures in the fecal microbiota of infants during the first months of life that were associated with a risk of being overweight, supporting the conclusion that the seeding of an “obesity-associated” microbiota in early life may transmit this phenotype to offspring (6). Similar data are emerging for a variety of other chronic illnesses including asthma, allergies, and autoimmunity, not just in childhood but into adulthood as well. This raises fundamental questions about our understanding of heritable phenotypes, blurring the lines between inherited familial disease risk and transmissible infection given that vertical transmission of microbiota loosely fulfills Koch's postulates (which form the criteria for establishing a causative relationship between microbial infection and disease).

With mounting and compelling data implicating the maternal microbiota of pregnancy and early life in health and disease, researchers have begun to uncover the biological mechanisms that allow the maternal and early-life microbiota to influence host physiology. As exemplified by studies of germ-free mice that are devoid of microorganisms, colonization by microbes in early life is critical for normal immune development, including immune cell production (comprising myeloid and lymphoid lineages), development of lymph nodes, antibody production, T cell polarization, and immune regulatory functions, as well as innate immunity and epithelial barrier integrity. Exposure of germ-free mice to commensal bacteria in the gut induces reprogramming of immune cell function that can reverse many of these immunological defects, but only (or much more effectively) if microbial exposure occurs at birth, not later in adulthood (10).

Given that the early-life microbiota is seeded by vertical transmission, the education of immunity in offspring is critically dependent on maternal commensals. Furthermore, this relationship among the maternal microbiota, early-life colonization, and immune development continues to coevolve throughout early life (see the figure). For example, breastfeeding serves as a continued source of maternal microbes, as well as supplying nutritional and antimicrobial factors that continuously shape the infant's gut microbiota (11). In support of this, an immunological “weaning reaction” in mice was recently observed at the stage of life when pups are weaned off breastmilk, consisting of a surge of microbiota-induced immune gene expression in the intestine together with the generation of immunosuppressive regulatory T cells (12). Perturbation of this critical developmental window caused long-lasting susceptibility to immunopathology later in life (12). Although it is clear that maternal microbes and other factors in breastmilk modulate microbiota composition and immunological function in offspring, the existence, magnitude, and importance of such a weaning reaction in humans has not been investigated. The immunological and developmental impact of the maternal and early-life microbiota extends well beyond the gut mucosal immune system, including emerging evidence for a microbiota-mediated gut–brain axis controlling neuroimmune development and susceptibility to neurodevelopmental disorders (e.g., autism) and psychiatric illnesses (13).

Effects of the maternal microbiota in pregnancy and early life

Through effects on early-life colonization, immune development, and neurodevelopment, the maternal microbiota regulates susceptibility to a number of childhood illnesses and can vertically transmit dysbiosis-mediated pathologies.


The maternal microbiota of pregnancy also appears to affect fetal development from afar. Prior to birth, the maternal microbiota is separated from the developing fetus by both physical barriers of the placenta and chorioamniotic sac and immunological barriers that are established at the maternal–fetal interface. Nevertheless, data from animal studies indicate that the composition and metabolic function of the maternal microbiota are reflected in fetal immune development. Deciphering the impact of the maternal microbiota during gestation on long-term childhood development has proven challenging owing to the inherent coupling of the gestational effects of the maternal microbiota with its role in establishing the early-life microbiota in offspring. For example, studies of mouse pups born to dams treated with antibiotics during gestation display abnormal T cell polarization and increased susceptibility to infections and autoimmunity later in life, but it is not clear whether this is solely a gestational effect versus a consequence of vertical transmission of a skewed microbiota to the pups (14, 15).

Fortunately, a clever model system has been developed to circumvent these limitations by making use of a transient colonization strategy to generate “gestation-only” colonization in mice. Transiently colonizing pregnant mice with a mutant strain of Escherichia coli that is unable to survive and replicate in vivo has enabled transient gastrointestinal colonization with these bacteria during gestation, followed by a return to germ-free status prior to delivery (16). With this model, fetal development occurs in the presence of a maternal microbiota, yet pups are born germ-free and never encounter microbes directly. This system revealed that gestational colonization programmed the development of gut mucosal immune cells in pups and induced vast changes in gastrointestinal gene expression, including increased epithelial antimicrobial peptide production. Outside of the gastrointestinal tract, pups born to gestationally colonized dams displayed reduced splenic cytokine production in response to bacterial endotoxin compared to germ-free counterparts. This proved that normal fetal immune development is critically dependent on long-distance signals received from the maternal microbiota, which functions to prepare the gut epithelium and mucosal immune system in the developing offspring for the onslaught of microbial exposure experienced at birth. Extending these findings to human biology will require further research and improved methodologies to define the gestational impact of the maternal microbiota on fetal development and childhood health and wellness.

Although much has been learned about the maternal microbiota in pregnancy and early life, many fundamental questions remain. It is unclear if this information can be used to prevent and treat childhood illnesses. To do so, are live microbial therapeutics needed, or will specific bacterial products or metabolites suffice (and if so, which ones)? Further research is needed to determine whether there is a “normal” maternal and early-life microbiota that should be universally strived for, or whether microbial therapeutics require individualization. In addition, what are the ethical considerations of therapeutic modification of an unborn fetus through manipulation of the maternal microbiota, such as issues of consent and the potential for unexpected adverse effects? To answer these (and other) outstanding questions, studies must move beyond observational correlations to define causal relationships, including the cellular and molecular mechanisms that mediate microbiota–host interactions in pregnancy and early life. This research must incorporate more than microbial community structure and instead aim to uncover the functional interactomes that exist between commensal microbes and the host. This represents an exciting challenge for scientists to refine the understanding of this fundamental aspect of human biology, and an opportunity to apply this science at the cutting edge of health care to improve maternal and child health.

References and Notes

View Abstract

Navigate This Article