Abstract
Early life determinants of the oral microbiota have not been thoroughly elucidated. We studied the association of birth and early childhood characteristics with oral microbiota composition using 16 S ribosomal RNA (rRNA) gene sequencing in a population-based Swedish cohort of 59 children sampled at 6, 12 and 24 months of age. Repeated-measurement regression models adjusted for potential confounders confirmed and expanded previous knowledge about the profound shift of oral microbiota composition in early life. These alterations included increased alpha diversity, decreased beta diversity and alteration of bacterial composition with changes in relative abundance of 14 of the 20 most common operational taxonomic units (OTUs). We also found that birth characteristics, breastfeeding and antibiotic use were associated with overall phyla distribution and/or with the relative abundance of specific OTUs. Further, we detected a novel link between morning salivary cortisol level, a physiological marker of neuroendocrine activity and stress, and overall phyla distribution as well as with decreased abundance of the most common OTU mapped to the Streptococcaceae family. In conclusion, a major part of the maturation of the oral microbiome occurs during the first two years of life, and this development may be influenced by early life circumstances.
Highlights
The oral microbiota represents a complex, dynamic, and heterogeneous bacterial ecosystem[1], and its composition has been discussed in relation to oral health[2,3] and to cardiovascular disease[4]
We observed a radical shift in the salivary microbiota composition from 6 months to 24 months of age, including a nearly threefold increase in alpha diversity as well as profound changes on all investigated taxonomic levels
We found a novel link between salivary cortisol and oral microbiota composition, encompassing both phyla distribution and lower relative abundance of the most common oral operational taxonomic units (OTUs) belonging to the Streptococcaceae family
Summary
As the microbiota analysis was performed in evening saliva samples, we only included morning salivary cortisol as anexposure in our regression analyses, as we wanted to reduce the risk of potential bias from a matrix effect. If an exposure was associated with distribution of bacterial phyla proportions, we carried out post-hoc tests, testing all phyla present in more than 75% of the samples individually using fractional logistic regression with cluster-robust standard errors. These post-hoc tests were not corrected for multiple comparisons. All women in the Born into Life cohort gave informed consent at enrolment during pregnancy, and both parents gave additional informed consent for the child
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
