In the target article, Hooks et al. critically review the current state of microbiota-gut-brain axis research in animal models and make specific suggestions on how to improve research in this area. However, Hooks et al. appear to have overlooked what might be considered one of the most promising avenues for moving research in this emerging field forward. Specifically, we would like to argue that the time is ripe to explore the role of the human microbiota in brain and cognitive development, especially during infancy (Kelsey et al. Reference Kelsey, Dreisbach, Alhusen and Grossmann2019).

From birth to age 3, the gut microbiome changes from a relatively sterile environment to a diverse ecosystem with thousands of species of bacteria, suggesting that this might represent a formative, and possibly sensitive, period in microbiota-gut-brain axis development (Borre et al. Reference Borre, O'Keeffe, Clarke, Stanton, Dinan and Cryan2014; Walker Reference Walker2013). The target article highlights initial support from animal models (e.g., Sudo et al. Reference Sudo, Chida, Aiba, Sonoda, Oyama, Yu, Kubo and Koga2004), showing that the timing of bacterial colonization plays an important role in the development of the gut-brain axis, yet it fails to acknowledge existing evidence from humans, which further supports the notion that early development during infancy may critically shape the microbiome-gut-brain axis. For example, both delivery and infant feeding methods, which have been shown to affect the gut microbiome composition in infants, have also been identified as risk factors for early emerging neurodevelopmental disorders such as autism spectrum disorder (Curran et al. Reference Curran, O'Neill, Cryan, Kenny, Dinan, Khashan and Kearney2015; Dominguez-Bello et al. Reference Dominguez-Bello, Costello, Contreras, Magris, Hidalgo, Fierer and Knight2010; Heikkilä & Saris Reference Heikkilä and Saris2003). These epidemiological findings provide indirect, correlational evidence for a microbiota-gut-brain axis link in early human development.

More direct evidence for such a link comes from a pioneering recent study by Carlson et al. (Reference Carlson, Xia, Azcarate-Peril, Goldman, Ahn, Styner, Thompson, Geng, Gilmore and Knickmeyer2018) in which fecal samples were collected from 89 typically developing infants and analyzed using 16S ribosomal RNA amplicon sequencing. In this study, the link between infant gut microbiome composition at 1 year of age and cognitive development (measured by the Mullen Scales of Early Learning) and brain development (measured using structural magnetic resonance imaging), at both 1 and 2 years of age were assessed. Carlson et al.’s (Reference Carlson, Xia, Azcarate-Peril, Goldman, Ahn, Styner, Thompson, Geng, Gilmore and Knickmeyer2018) analysis revealed that cognitive development scores differed significantly between infants assigned into one of three gut microbiome taxonomic groups, as identified by cluster analysis. This study also reports some specific structural brain differences linked to the microbiome composition. However, it should be noted that the majority of structural brain measures, such as intracranial volume, total white or gray matter, total cerebrospinal fluid, or lateral ventricle volume, did not reveal any differences between infants for the three bacterial composition groups. Moreover, from these data, it is still unclear how the small volume differences found in specific brain areas are related to infant brain and cognitive function. Contrary to what is known from adults where higher microbial diversity has typically been shown to be predictive of positive health outcomes (Abrahamsson et al. Reference Abrahamsson, Jakobsson, Andersson, Björkstén, Engstrand and Jenmalm2014; Kostic et al. Reference Kostic, Gevers, Siljander, Vatanen, Hyötyläinen, Hämäläinen, Peet, Tillmann, Pöhö, Mattila, Lähdesmäki, Franzosa, Vaarala, de Goffau, Harmsen, Ilonen, Virtanen, Clish, Orešič, Huttenhower, Knip and Xavier2015), Carlson et al. (Reference Carlson, Xia, Azcarate-Peril, Goldman, Ahn, Styner, Thompson, Geng, Gilmore and Knickmeyer2018) showed that increased microbial alpha diversity was associated with lower cognitive performance in infancy. Based on this discrepancy, Carlson et al. (Reference Carlson, Xia, Azcarate-Peril, Goldman, Ahn, Styner, Thompson, Geng, Gilmore and Knickmeyer2018) suggest that microbial diversity may affect cognitive functions differently in infancy than later in development. In any case, the study by Carlson et al. (Reference Carlson, Xia, Azcarate-Peril, Goldman, Ahn, Styner, Thompson, Geng, Gilmore and Knickmeyer2018) as a first of its kind sheds new light on how individual differences in brain and cognitive development during infancy emerge in the context of the developing microbiome-gut-brain axis. Collectively, this points to the importance of developmental research, which systematically maps associations between microbial characteristics and brain and cognitive development across the entire human life span.

Related to taking a human developmental perspective, another potentially overlooked research approach is underscoring the use of developmentally informed humanized mouse models in order to create more translatable research. In the target article, authors make a poignant argument that there are inherent issues when one tries to make inferences about human mental health disorders from studies with animal models. The authors suggest that this area of research often uses language that overextends the implications of germ-free mouse models and rodent behavioral tests to human mental health. However, they fail to mention an alternative methodological strategy to addressing the issue of translatability, which is by creating humanized mouse models (for a review, see Walsh et al. Reference Walsh, Kenney, Jangalwe, Aryee, Greiner, Brehm and Shultz2017). Specifically, fecal samples from humans can be taken from clinically relevant populations (with or without mental health issues) at different points during development (from newborns to aging populations) and transplanted into animals – thus creating developmentally informed animal models that allow for a more mechanistic study of the microbiome-gut-brain axis.

Finally, we would like to argue that the implications for research on the early development of the microbiome-gut-brain axis in humans extend well beyond the somewhat overemphasized field of probiotic research. Specifically, in the context of infant development, research in this field has potentially major implications for delivery and neonatal care procedures. For example, medical facilities have recently started to examine the health benefits of “seeding” procedures, whereby infants delivered via C-section are wiped with maternal vaginal swabs, with the hope of colonizing infants with more diverse groups of bacteria. Moreover, the benefits of breastfeeding on infant brain and cognitive development have been widely studied and documented (Krol & Grossmann Reference Krol and Grossmann2018). However, the gut microbiome has been largely ignored as a potential contributor to the positive effects of breastfeeding on infant and child development. Therefore, recognizing the need for incorporating a microbiome perspective in delivery and breastfeeding research with infants might help inform clinical practice.

Taken together, this commentary is intended to emphasize the importance of looking at the microbiota-gut-brain axis from a human developmental perspective with a specific focus on infancy. In addition, this commentary is meant to encourage the use of humanized animal models to tackle translatability issues and realize implications of this work, which extend well beyond probiotic administration. Overall, the hope is to complement the target article by inspiring the bold research programs needed to systematically examine the microbiome's role in early human brain and cognitive development.