Prenatal Microbial Exposure

Based on the Barker hypothesis for the “developmental origins of adult disease,” the prenatal environment has been implicated in the development of later disease, with examples such as low birth weight or prenatal undernutrition being linked to adult weight, type 2 diabetes, hypertension, and coronary artery disease. Maternal microbial exposure during gestation may have some direct effects on the fetus, and the infant’s immune development and long-term health.

Direct bacterial exposure of the fetus through amniotic fluid, which is typically sterile, may have deleterious effects. Bacterial ribosomal DNA and bacteria from Leptotrichia spp and other related species have been identified in amniotic fluid of women in preterm labor, and studies suggest that infant prematurity is positively associated with bacterial load, even in the absence of ruptured membranes. Although the clinical significance of microbial identification in amniotic fluid has not been fully identified, investigators have suggested that microbes within amniotic fluid swallowed by the fetus could translocate through the intestinal epithelium and elicit inflammation, and potentially stimulate uterine contraction and premature delivery.

On the other hand, maternal exposure to microbes during pregnancy seems to play a role in postnatal immune functioning, particularly regarding allergic disease. Maternal prenatal exposure to household pets (cats and dogs) has been associated with lower cord blood IgE concentrations and increased number and function of cord blood T-regulatory cells, and a lower incidence of allergic disease in children. Mothers exposed to farms and farm animals during pregnancy were also less likely to have children who developed allergies and asthma. The degree and diversity of maternal exposure to microbes may also have an impact. Higher levels of innate immune components, Toll-like receptors (TLR2, TLR4, and CD14), have been identified in children whose mothers were exposed to stables during the prenatal period, and for every extra farm-associated animal species a mother encountered, expression levels increased by up to 1.16-fold.

Evidence shows the potential of small, yet immunogenic, allergen fragments to cross the placenta and prime naïve fetal T cells, resulting in a form of natural immunity during gestation. These allergens, in concert with maternal immunologic responses to microbial exposure, seem to shape the fetal immunologic environment in a way that can affect postnatal predisposition to disease.

Mode of Birth

During vaginal birth, microbes from the birth canal, the perineum, and the mother’s skin constitute the first bacterial inocula for gastrointestinal luminal colonization. Thus, the infant’s early microbiota profile is similar to the mother’s vaginal and fecal microbes. Although individual variation exists, facultative anaerobic E coli and Streptococcus spp tend to dominate in infant stool in the first 2 to 3 days after birth. However, within a few days, anaerobic genera, such as bifidobacteria and bacteroides (which have been associated with beneficial effects, including down-regulation of inflammatory responses), become quantitatively more important in the infant stool of vaginally born infants.

In contrast, infants born by cesarean section (CS) show delayed colonization and lack of microbial diversity. Differences in composition of microbiota during the first days, months, and years of life have been reported for infants delivered by CS compared with infants born vaginally. Often, the early microbiota of infants delivered by CS consists of species found within the bacterial hospital environment, including reduced levels of strict anaerobes such as bifidobacteria. Compared with vaginally born infants, those born by CS tend to harbor higher stool quantities and/or increased prevalence of Staphylococcus spp, Streptococcus spp, C difficile, klebsiella, and enterobacteria, and a reduced prevalence of bacteroides, and reduced or delayed colonization with bifidobacteria and lactobacilli. These altered patterns have been documented up to 7 years of age.

Adequate data are not available to compare microbiota of infants delivered by CS at the time of membrane rupture and those with early membrane rupture who were allowed some exposure to vaginal flora. Investigators have suggested that physical passage of the infant through the birth canal may be more important than the presence or duration of rupture of membranes in determining early microbial composition.

In summary, infants born by CS experience significant deviation from the normal pattern of microbiota acquisition, and develop gastrointestinal microbial profiles distinctly different from those of infants born vaginally. This “aberrant” type of intestinal microbiota has been implicated in altered immunologic host responses that could lead to immune-related conditions in later life. CS delivery has been identified as an independent risk factor for the development of several conditions, including allergic disease, including asthma, and gastroenteritis in later childhood. Meta-analysis findings from 23 studies on the association between mode of birth and allergic disease showed a 20% increase in the development of asthma in children delivered by CS compared with those delivered vaginally. A more than 7-fold increase in risk of allergy has been reported for children born to mothers with a history of allergy who were also delivered by CS. These relationships remained significant after adjusting for confounding variables, including short-term breast-feeding.

Of even greater interest is the increasing identification of CS as an independent risk factor of immune-related conditions other than allergic disorders. Various studies have investigated a relationship between CS delivery and type 1 diabetes. A recent meta-analysis of these studies showed a 20% increase in the risk of childhood-onset type 1 diabetes associated with CS delivery that was not explained by known confounders. In a large retrospective, multicenter, case-control study, a significantly enhanced likelihood of being born by CS was found in children with celiac disease compared with control subjects. In this cohort CS delivery was not found to be associated with Crohn disease or ulcerative colitis.

Finally, recent studies suggest that obesity and metabolic syndrome may be associated with changes in microbiota. A prospective cohort study showed that infants delivered by CS had 2-fold higher odds of obesity at age 3 years, even after adjusting for maternal body mass index, birthweight, and other confounding variables. Although causality and potential mechanisms remain to be elucidated, increasing evidence shows a relationship between the establishment of microbiota in early life and subsequent phenotypic manifestations in the host, including changes in adiposity and related metabolic alterations.

The increasing recognition of CS as a risk factor for chronic conditions that manifest themselves far beyond the perinatal period should foster increased awareness of these risks, and serve as additional argument against non–medically indicated CS.

Dietary Intake

Diet is one of the most important determinants of microbial diversity in the gut. In the full-term breast-fed infant, anaerobic bacteria from the genus Bifidobacterium begin to colonize the gut within the first week of life. Bifidobacteria can constitute more than 60% of the fecal bacteria in breast-fed infants within 6 days of nursing, and up to 72% by the third week of exclusive breast-feeding. Factors likely responsible for a strong dominance of bifidobacteria in the stool of breast-fed infants include the presence of multiple Bifidobacterium spp in maternal milk that provide a constant inocula to the infant, and several breast milk components, including galacto-oligosaccharides, which selectively foster the growth of bifidobacteria in the gut.

Compared with breast-fed infants, formula-fed babies are less frequently colonized with bifidobacteria, tend to have bifidobacteria in lower numbers, and are more often colonized with potentially pathogenic species of enterococci, coliforms, and clostridia. Specifically, populations of E coli , C difficile , and bacteroides have been rather consistently reported as being more prevalent or reaching higher counts in stool from formula-fed infants.

Although limited studies have reported no significant difference in the numbers of bifidobacteria in stools of exclusively formula- or breast-fed infants, Klaassens and colleagues identified that the functional gene expression and type of bifidobacteria present in infant stool was shown to differ with mode of feeding. Moreover, the widespread inclusion of lactose and compounds similar to bifidogenic human milk oligosaccharides, such as galacto-oligosaccharides, in commercial formulas may account for some less-than-expected differences in stool microbiota of formula-fed infants.

By approximately 2 years of age, when children are consuming an adult-type diet, the gut community begins to resemble that of an adult-like microbiota. Once established, both the microbiota and caloric contribution of various food groups in late infancy remain reasonably stable. However, clinically induced variations in dietary macronutrient contribution have been associated with changes in microbiota. Increased numbers of Firmicutes and reduced numbers of Bacteroidetes and bifidobacteria have been documented with high-fat feeding in both animal and human studies, although long-term follow-up studies are not available to determine if these changes persist.

Antibiotic Exposure

Antibiotic use can significantly alter the composition of intestinal microbiota, and may do so in specific ways in infants. Dramatic decreases in microbial diversity with antibiotic administration during the first year of life have been reported in term infants. Administration of antibiotics (primarily cephalosporins or oral amoxicillin) in the first months of life resulted in reduced numbers of fecal bifidobacteria and bacteroides and overgrowth of C difficile . In premature infants, the duration of antibiotic treatment in the first month of life has been shown to correlate with decreased bacterial diversity. Enterococci populations were increased 1 month after a 4-day treatment of broad-spectrum antibiotics in neonates.

Unlike adults, who may return quickly to pretreatment microbiota after cessation of antibiotic treatment, the effect of antibiotics on infant’s microbiota may affect immune homeostasis and create immediate susceptibility to disease with long-lasting consequences. Antibiotic exposure during the first year of life has been identified as an independent risk factor for wheezing, and meta-analyses identify it as a risk factor for childhood asthma. Early antibiotic exposure has been associated with increased risk of childhood atopy, and increased frequency of antibiotic use also correlated with atopic risk. These findings underscore the importance of the initial period of host bacterial experience, and a critical window in which aberrations in colonization patterns may induce long-term changes in microbiota and in immunologic responses, with long-term consequences.

Other Factors

The immediate environment in early life can also affect microbiota composition. The gut is colonized with a only a small number of bacterial species in hospitalized, premature infants; lactobacilli and bifidobacteria are rarely identified in microbiota. Hospitalization of neonates after birth for as little as 4 days has been associated with decreased prevalence of bifidobacteria and higher C difficile colonization rates.

Finally, the role of the individual, the unique host harboring this complex ecosystem of bacteria, cannot be underestimated. Individual differences in the luminal environment provided to the microbiota can also affect its composition. Differences in factors such as oxygen tension and lumen redox potential, pH, the composition of digestive enzymes, biliary secretions, mucus and mucin, and IgA production are all specific and phenotypically unique to each individual host, and these variations can modulate the microbiota in ways that are only beginning to be understood.

Atopic Diseases

Much of the evidence for the link between early microbial populations and subsequent development of immune disorders come from studies of atopic disease, and the evidence is strong for a link between early gastrointestinal colonization and subsequent development of allergic manifestations, such as eczema and asthma. Evaluation of approximately 1000 stool samples of 1-month-old infants identified a high prevalence of E coli associated with later development of eczema, and those with a high count of C difficile were associated with a higher risk of eczema, recurrent wheeze, allergic sensitization, and atopic dermatitis. In addition, allergic infants and children were found to have less colonization with lactobacilli and bifidobacteria, and the prevalence of colonization with bifidobacteria in infants who developed allergy during the first 2 years of life was less than for those that did not develop atopy. In addition to decreased bifidobacteria, allergic infants were colonized less often with Bacteroides spp and more often with Staphylococcus aureus. Furthermore, a reduced ratio of bifidobacteria to clostridium in early gut microbiota has been reported to precede allergic disease.

Overall, a more diverse early gut microbiota is more common among nonallergic infants than in those with allergy, yet whether low diversity of the gut microbiota in infancy is more important than the prevalence of specific bacterial taxa in the development of allergic disease has been a matter of debate. Infants with sensitization or eczema were reported to have fewer microbial peaks/bands than healthy infants, without regard to specific microbes. High-throughput 16S-based molecular microbiology recently confirmed these previous findings. More recent results indicate that low intestinal microbial diversity during first month of life was associated with subsequent atopic eczema and may be more relevant than any specific bacteria. One proposed rationale for the importance of microbial diversity is based on the assumption that the gut immune system reacts to exposure to new bacterial antigens, and these repeated exposures of various microbes would enhance the development of immune regulation.

In summary, differences in microbiota among infants developing allergy and those who do not suggest that factors in early life that alter gut microbial composition may be a determinant in the later development of this immunologic disease. In fact, cesarean birth, antibiotic use, and formula-feeding (rather than breast-feeding), all of which are associated with the altered patterns of microbiota, are also independent risk factors for allergic conditions.

Obesity and Metabolic Syndrome

Low-grade systemic inflammation has been associated with overweight, obesity, and disorders directly linked to adiposity, such as insulin resistance and type 2 diabetes. Microbiota has been suggested to be a contributing factor in the development of some of these metabolic disorders. Several studies have shown that, compared with lean individuals, those who are obese harbor a greater concentration of Firmicutes and have an exaggerated shift in the ratio of Firmicutes to Bacteroidetes (from 3:1 to 35:1), 2 of the major phyla present in the adult human gastrointestinal tract. This altered bacterial quantity has been implicated in obesity development through its effects on increasing energy harvest from food and its interactions with epithelial and endocrine cells that promote insulin resistance and inflammation and increase adipocyte generation. Thus, microbiota, influenced by acquisition and composition in early life, may play a significant role in the development of overweight or obesity.

One recent prospective trial of 138 infants identified (by culture on selective media) that a microbiota with high Bacteroides fragilis and low staphylococcus concentrations in infants between the age of 3 weeks and 1 year was associated with a higher risk of obesity during preschool age. A separate longitudinal study of microbiota composition during infancy (at 6 and 12 months) and 7 years later examined 25 children who became overweight/obese by age 7 and 24 children who remained normal weight. Results from FISH analysis and quantitative PCR showed that the amount of fecal bifidobacteria was greater and the concentration of S aureus lower in children remaining normal weight compared with those who later became overweight. The authors proposed that protection from obesity noted with higher bifidobacteria may be partly mediated by its potential anti-inflammatory effects, whereas S aureus may act as a trigger of low-grade inflammation, contributing to difficulty with weight management. A third study that examined microbiota composition during infancy (3 weeks and 3 months of age) through FISH analyses identified a trend toward reduced bifidobacteria counts in 3-month fecal samples and overweight status in children at age 10 years, compared with their normal-weight counterparts.

Mechanistic studies and a more comprehensive understanding of obesity and metabolic disorders in relation to gut microbiota are necessary. Nevertheless, the fact that by 2 years of age the microbiota reaches a stable compositional state, and that adiposity measures starting at 2 years of age track into later childhood, provides additional impetus for research in this area.

Necrotizing Enterocolitis

Premature infants are a population at particular risk of developing an aberrant microbiota composition. They are often born by CS, most are administered antibiotics, they are infrequently breast-fed, and their first microbial surrounding is the hospital environment. As would be expected, premature infants have delayed patterns of colonization, decreased microbial diversity, increased numbers of potentially pathogenic bacteria, and decreased colonization with bifidobacteria. These deviations are often identified in premature infants who develop necrotizing enterocolitis (NEC). Specific strains of klebsiella, enterobacteria, or E coli in microbiota have been shown to precede the development of NEC, and fecal colonization with enterococcus and Candida albicans has been identified more frequently in symptomatic infants than controls.

Infantile Colic

Potential mechanisms implicated in the occurrence of bouts of excessive and inconsolable crying in infants include behavioral, digestive, and gastrointestinal motility factors. Recently, differences in gut microbiota have been reported between infants with symptoms of colic and those without. Infants with colic seem to harbor more gas-producing bacteria (eg, E coli ) and be less frequently colonized by non–gas producing microbes (eg, lactobacilli and bifidobacteria). Other recent studies identified that colicky infants were more likely to have a restricted bacterial diversity and were more frequently colonized with coliforms such as Klebsiella spp, Enterococcus spp , and E coli than age-matched controls ; these findings remained when differences in exposure to antibiotics and types of feeding were considered. Whether abnormal microbiota is the cause of infantile colic or the result of intestinal luminal factors such as inflammation has yet to be determined. Similarly, whether an association exists between intestinal bacteria and other functional disorders, such as functional abdominal pain and irritable bowel syndrome, awaits additional investigation.

Crohn Disease

A decreased quantity and biodiversity of bacteria within the Firmicutes phylum has been repeatedly observed in patients with Crohn disease (CD), and further evidence of this comes from a recent study with twins designed to consider genetic influences. When comparing discordant and concordant twin pairs, regardless of disease state, the intestinal microbial composition showed a dramatically lower abundance of Faecalibacterium prausnitzii and increased abundance of E coli in twins with CD, compared with healthy co-twins and those with CD localized in the colon. The reduction in F prausnitzii , a major member of the Firmicutes phylum, which has anti-inflammatory properties, was associated with an increased risk of recurrence of ileal CD. Using a variety of assessment methodologies, decreased bifidobacteria and lactobacilli have been found in adults and children with CD.

In summary, several pathologic conditions are related to alterations in microbiota composition when compared with healthy individuals. Although most of these observations of various dysbiosis linked to disease do not prove causality, they indicate a close relationship between intestinal microbiota and numerous gut-barrier, immune, and possibly also metabolic functions of the host, which argues strongly for a major role of gastrointestinal microbes in the development and modulation of these disorders. Additional evidence of this is derived from transplantation experiments in which the altered microbiota from a diseased animal was provided to a germ-free healthy recipient; several disease phenotypes could be transferred by the microbiota. These include excess adiposity, metabolic syndrome, and colitis. In humans, fecal transplantation with donor fecal flora, have shown some success in treatment of Crohn’s disease, C difficile –associated diarrhea, ulcerative colitis, and chronic constipation.

Finally, the possibility of beneficially affecting the host’s health through modulating the composition of the intestinal bacterial composition via ingestion of bacteria, the concept of probiotics, has resulted in an explosion of research over the past 2 decades. The use of specific bacteria, particularly species from the genus Bifidobacterium and Lactobacillus , has been shown with varying but increasing levels of documentation to have both preventive and therapeutic benefits in several conditions, including allergic conditions, inflammatory gastrointestinal disorders, NEC, and functional gastrointestinal disorders, including infantile colic, all of which are associated to alterations in the microbiota. Although this topic is beyond the scope of this review, the effects of orally administered bacteria, which change the gastrointestinal ecosystem and improve those same conditions associated to altered patterns of gastrointestinal microbiota, provide additional evidence of a direct causal link between the microbiota and host health.