Elevated Fecal pH Indicates a Profound Change in the Breastfed Infant Gut Microbiome Due to Reduction of Bifidobacterium over the Past Century

1There is clear evidence that the infant gut microbiome has important long-term health implications, but changing the gut microbiome is challenging. We recently observed changes in fecal pH resulting from Bifidobacterium infantis EVC001 colonization owing to this bacterium’s selective and acidic fermentation of human milk oligosaccharides (HMOs), which was associated with a reduction in taxa that are signatures of dysbiosis. Although remodeling of the gut microbiome in breastfed infants fed B. infantis EVC001 improved gut function and ecosystem productivity, questions remain about whether differences in Bifidobacterium abundance and species between resource-rich and resource-poor countries are due to host genetics, geography, medical interventions, and/or demographics. Here, we show evidence for an increase in infant fecal pH over the past century, corresponding to an observed reduction of Bifidobacterium, the keystone infant gut symbiont. This may have implications for epidemic human immunological dysfunctions as perturbations in microbiota composition can lead to chronic inflammation and immune-mediated diseases.

2In 1913, Logan described the breastfed infant gut microbiome as being an “almost pure culture” of a Gram-positive, acidiphilic “Bacillus bifidus” (Bifidobacterium). This early microscopic characterization of diet-dependent infant microbiomes is in stark contrast to modern reports from resource-rich countries of unstable and highly diverse microbiomes. Recent comparisons of the infant gut microbiome from genetically similar but demographically diverse backgrounds indicated that Bifidobacterium was more abundant among infants from resource-poor locations, consistent with infants in sub-Saharan Africa and South Asia. These differences are also notable at the species level, in that the Bifidobacterium in the feces of infants in Gambia and Bangladesh were shown to be predominantly Bifidobacterium longum subsp. infantis (B. infantis), whereas the Bifidobacterium species in stool samples from infants in the United States and Europe consisted predominantly of B. breve and B. longum subsp. longum (B. longum). Substantial differences in Bifidobacterium composition and abundance among populations have led to questions as to whether medical interventions (e.g., caesarean section, antibiotic use) and formula feeding, or geographic and genetic differences alone, results in these differences.

3Recently, we found that breastfed infants fed B. infantis EVC001 developed a stable population of this strain and experienced substantial changes in intestinal biochemistry. Notably, fermentation of HMOs resulted in the increased production of lactate and acetate, which was markedly lower in infants who lacked populations of Bifidobacterium or were colonized by other Bifidobacterium species. This was concurrent with significantly higher fecal excretion of HMOs than that of infants fed B. infantis. Using data published by Frese et al., we compared fecal pH measurements with bacterial taxa by using a Spearman correlation. Importantly, only one family was significantly associated with reduced fecal pH, i.e., Bifidobacteriaceae (P < 0.0004), indicating that while other bacteria can consume HMOs (e.g., Bacteroidaceae), only members of the family Bifidobacteriaceae convert them to acidic end products with a meaningful effect on fecal pH (Fig. 1). This corroborates previous findings linking infant fecal pH to Bifidobacteriaceae abundance. This is a critical connection because although other bacteria (e.g., Lactobacillus, Clostridiaceae, Lachnospiraceae, and Ruminococcaceae) may produce organic acids during fermentation (e.g., lactate, acetate, butyrate, propionate), they were not significantly associated with the acidic fecal pH in breastfed infants. Infants colonized by B. infantis EVC001 had negligible levels of HMOs in their feces and an average fecal pH of 5.15, whereas infants lacking B. infantis had 10-fold higher levels of HMOs in their feces and a fecal pH of 5.97. Further, quantitative PCR confirmed the association of low fecal pH with an increased abundance of Bifidobacterium, in agreement with another study.

4Early 1900s reports suggest a rapid reduction in the fecal pH of breastfed infants during the first week after birth. Gyorgy and others identified a “bifidus factor,” whose abundance contributed to this reduction in fecal pH and an increase in Bifidobacterium in infant feces. This “bifidus factor” (now collectively described as HMOs), is selectively consumed by infant-associated Bifidobacterium; therefore, pH may be a reliable proxy of the breastfed infant gut microbiome. Infant fecal pH reported over the past century is independent of microbiological methodologies (e.g., microscopic examination versus 16S rRNA gene sequencing); thus, we speculated that historical reports of fecal pH could be used as an indirect measure of Bifidobacterium abundance.

5Fourteen peer-reviewed studies published between 1926 and 2017 and reporting 312 measurements from healthy, breastfed infants were found and included. A least-squares linear regression model revealed a strong positive trend with a high association between the publication year and fecal pH (slope = 0.014, adjusted r2 = 0.61; Fig. 2). These data suggest that the mean fecal pH of breastfed infants has increased from about 5.0 in 1926 to 6.5 in recent years (Table 1). Given our previous finding linking fecal pH to Bifidobacterium abundance and reported differences in Bifidobacterium abundance across populations today, this longitudinal change is consistent with a generational loss of Bifidobacterium in developed countries, most notably among infants born after 1980.

6The absence of Bifidobacterium as a keystone symbiont in infants may explain the increase in fecal pH and can be linked to unintended historical and generational consequences of certain interventions that have otherwise significantly improved infant and maternal health. First, a rapid increase in the use of human milk replacers (e.g., evaporated milk and infant formula), which lack the bacterial selectivity of human milk, beginning in the 1920s may have resulted in the inability to foster high levels of specialized infant-associated Bifidobacterium in the infant gut among nonbreastfed infants. This may also explain why B. infantis, which is highly specialized for the consumption of HMOs, is now exceptionally rare among infants in the United States and Europe, whereas B. longum and B. breve, which can access mucin glycans and plant carbohydrates, remain relatively abundant. Second, increased caesarean section delivery since the 1980s further limits the natural fecal-oral transfer of Bifidobacterium from mother to infant associated with vaginal delivery. Third, antibiotic use has become increasingly common during labor and many infant-associated species of bifidobacteria are sensitive to antibiotics. For example, the use of antibiotics to prevent the transmission of group B Streptococcus during delivery and the use of caesarean section as the mode of delivery are both critically important interventions in public health but can alter the acquisition of gut microbes by the infant that begins at birth. Together, these barriers may have played a role in the loss of Bifidobacterium over time and across generations, which is reflected in a higher fecal pH.

7There is clear evidence that the infant gut microbiome has important long-term health implications, and perturbations of the microbiome composition may lead to chronic inflammation and immune-mediated diseases. These data highlight an increase in infant intestinal dysbiosis. Thus, the loss of Bifidobacterium and the profound change in the gut environment, as measured by fecal pH, present a compelling explanation for the increased incidence of allergic and autoimmune diseases observed in resource-rich nations. Longitudinal analyses studies comparing the incidence of autoimmune disorders with restored Bifidobacterium populations in the infant gut microbiome are essential to establish the role of Bifidobacterium in early immune development in the infant gut.