Chair:
Dr Sarah Vogel – Boston University, Boston, USA
Symposium Presenters:
- Sarah Vogel, Boston University, Boston, USA
- Patricia Brennan, Emory University, Atlanta, USA
- Naomi Gancz, University of California, Los Angeles, USA
- Harikesh Dubey, Michigan State University, East Lansing, USA
The gut microbiome has been trans-diagnostically linked with psychopathology in adults, and a wave of recent research has suggested that differences in the diversity and composition of the gut microbiome in early life may be associated with differential risk for the development of psychopathology. The gut microbiome undergoes a critical period of development in early life, overlapping with sensitive periods in other key physiological and stress response systems, making questions about connections between the gut microbiome and psychopathology particularly important to consider from a developmental perspective. In this symposium, we explore the role of the gut microbiome in developmental psychopathology, the state of the literature in this growing line of inquiry, and identify areas for future research. In the first presentation, Dr. Sarah Vogel will present longitudinal associations between development of the gut microbiome throughout infancy and behavioral measures of fearfulness and negative affectivity in early childhood. Next, Dr. Patricia Brennan will integrate perspectives from the social determinants of health and developmental origins of health and disease (DOHaD) frameworks to discuss connections between prenatal stress, the infant gut microbiome, and child behavior problems in two samples of African American infants and their mothers. Naomi Gancz will then discuss connections between the gut microbiome and internalizing symptoms across childhood and the role of childhood adversity in moderating gut-behavior associations. Finally, Dr. Harikesh Dubey will present some experimental work examining the effects of fecal transplant procedures in mice and implications for anxiolytic behavior, with an emphasis on neural mechanisms, including myelination and dendritic complexity. This symposium will integrate research perspectives from across disciplines, species, domains, and developmental timing to contextualize the state of our scientific understanding of the role of the gut microbiome in developmental psychopathology and inspire discussion about the potential future directions for research and intervention.
Abstract for Presentation 1
EXAMINING GUT MICROBIOME DEVELOPMENT ACROSS INFANCY AND FEAR BEHAVIORS IN EARLY CHILDHOOD
Sarah C. Vogel1, Nicolas Murgueitio2, Nicole Huth1, Kathy Sem1, Rebecca Knickmeyer3, Roger Mills-Koonce2, Cathi Propper2, and Nicholas J. Wagner1
- Boston University, Boston, MA
- University of North Carolina at Chapel Hill, Chapel Hill, NC
- Michigan State University, East Lansing, MI
The gut microbiome has been implicated in a range of psychiatric diagnoses in adults, including anxiety and depression. In young children, a few studies have documented associations between the gut microbiome and early indicators of risk for the development of anxiety, including non-social fear, functional brain connectivity, and internalizing problems. Most existing studies in this area have been cross-sectional, and the few studies that include repeated measurement of the gut microbiome have yet to focus on change in the gut microbiome across development as it is associated with behavioral outcomes. Addressing this research gap is important because the gut microbiome undergoes a sensitive period of development in the first three to four years of life, during which there are rapid increases in the diversity of gut bacteria and large shifts in the composition of the gut microbiome.
In this preregistered study of 64 infants and their caregivers, we use latent change scores to characterize change from 2 weeks to 18 months in gut microbiome alpha diversity, a global metric of gut microbiome diversity, and the relative abundance of specific bacterial taxa in the samples. Informed by research linking Research Domain Criteria constructs (RDoC) to anxiety risk, we focus on negative valence systems as our measure of anxiety risk at 30 months, measured via a researcher report of infant affect in the Infant Behavior Record following a study visit.
Our results reveal intriguing associations. We found that higher 2-week alpha diversity is significantly associated with more fearfulness and negative infant affect at 30 months (β=-0.36, p=.01). We also identified several genera of bacteria at 2 weeks whose abundances were associated with infant affect at 30 months, including Veillonella, Streptococcus, and Enterocloster. These findings were robust to corrections for multiple comparisons and inclusion of various covariates. Interestingly, we did not find associations between markers of growth in the gut microbiome across time or markers of the gut microbiome at 18 months and infant affect at 30 months, suggesting a potential effect of developmental timing on connections between the gut microbiome and anxiety risk. The implications of these findings for future research and intervention strategies will be discussed in detail.
Abstract for Presentation 2
MATERNAL PERINATAL STRESS, BREASTFEEDING, AND CHILD NEURODEVELOPMENT: ASSOCIATIONS WITH THE INFANT MICROBIOME IN AN AFRICAN AMERICAN SAMPLE
Patricia A. Brennan1, Brooke G. McKenna1, Anne L. Dunlop2 and Elizabeth J. Corwin3
1 Department of Psychology, Emory University, Atlanta, GA, USA
2 Department of Gynecology and Obstetrics, School of Medicine, Emory University, Atlanta, GA, USA
3 School of Nursing, Columbia University, New York, NY, USA
Health disparities begin in utero and continue after birth. Compared to White infants, African American infants are more likely to evidence neurocognitive delays, a disparity that persists or worsens across development, even when controlling for socioeconomic status. Increased chronic stress exposure during pregnancy and a differential susceptibility to maternal perinatal stress are hypothesized to significantly contribute to these developmental disparities. Rates of breastfeeding are also lower in African Americans, potentially reducing protective nutritive influences on children’s outcomes. Recently, the gut microbiome has been suggested as an important link between stress exposure and neurodevelopment, but microbiome studies focused on African American infants are scarce.
Participants in this study included two overlapping samples. The first sample consisted of 79 African American mothers (aged 18-35) and their infants. During pregnancy and infancy follow ups, mothers in this sample reported on life stressors and perceived stress symptoms. Breastfeeding status was ascertained, and infant stool was collected at one month and three months of age. The stool was later batch assayed using 16S sequencing methods to assess levels of gut microbiome taxa. The second sample consisted of 114 African American mothers (aged 18-35) and their infant and toddler children. Infant stool from 6-24 months of age was assayed to assess gut microbiome taxa levels, and mothers reported children’s behavioral outcomes at two and three years on the Child Behavior Checklist. In the first sample, regression models were used to examine effects of maternal stress on Lactobacillus and Bifidobacterium taxa, and to test breastfeeding as a potential moderator. In the second sample regression models were used to test associations between Lactobacillus and Bifidobacterium taxa and child behavior problems. All analyses adjusted for maternal education, maternal age, maternal prenatal BMI, parity, mode of delivery, offspring sex, visit time point, and offspring age at stool collection.
In the first sample, higher levels of prenatal stressful life events were associated with lower levels of infant gut Bifidobacterium (p<.05), and breastfeeding status interacted with maternal reports of perceived stress during pregnancy and the postpartum to predict Bifidobacterium levels (p<.05) such that negative associations between maternal perceived stress and Bifidobacterium were attenuated in cases where infants were currently being breastfed. Contrary to our hypothesis, maternal postpartum perceived stress levels were associated with higher rather than lower levels of Lactobacillus (p<.001). In our second sample lower levels of Bifidobacterium and higher levels of Lactobacillus predicted increased behavior problems in children (p<.05).
Results reveal associations between maternal stress, the infant gut microbiome and child neurodevelopment, some of which suggest a protective role for infant breastfeeding.
Abstract for Presentation 3
ADVERSITY MODERATES THE QUADRATIC DISTRIBUTION OF GUT MICROBIOME FEATURES ACROSS AGE – IMPLICATIONS FOR INTERNALIZING SYMPTOMS
Gancz, N.N.1, Silvers, J.A.1, Choy, T.2, VanTieghem, M.3, Tottenham, N.3, Callaghan, B.L.1
1University of California, Los Angeles
2University of California Riverside
3Columbia University
Youth exposed to caregiving adversities (CA) are at elevated risk for internalizing symptoms, but much remains unknown about the mechanisms underlying this risk. Gut microbiome features, especially short-chain fatty acid (SCFA) and aromatic amino acid (AAA) metabolism, are thought to be causally linked to internalizing symptoms. However, much remains unknown about how the abundance of such microbiome features varies across age in typically developing youth and, even more so, in youth exposed to CA. Given that childhood and adolescence encompass periods of rapid neurodevelopment and internalizing symptom onset, this information is of utmost utility. Using a systematic model-building approach, we examined the associations between CA and taxonomic gut microbiome composition in a cross-sectional sample of N = 102 youth aged 1-18. We then tested whether CA-linked microbiome features were associated with internalizing symptoms, measured via parent reports on the Child Behavior Checklist. Unexpectedly, we found that CA moderated a quadratic distribution of several bacterial genera across age. In order to elucidate functional implications of these taxonomic findings, we estimated SCFA and AAA metabolic pathways, which were also significantly associated with the interaction of CA and the quadratic term of age. As expected, several of the taxonomic and estimated functional features that were associated with CA or its interaction with age were also associated with internalizing symptoms. Our results demonstrate that the distribution of gut microbiome features across age is, in multiple cases, not only curvilinear, but also dependent on adversity exposure. Moreover, several of these microbiome features can also explain variance in internalizing symptoms. In future work, longitudinal data and functional microbiome sequencing will provide increased granularity and understanding of changes in microbiome distribution over time.
Abstract for Presentation 4
VARIABILITY IN HUMAN INFANT MICROBIOMES HAS ALTERED THE DENDRITIC MORPHOLOGY IN GERM-FREE MICE
Harikesh Dubey, Postdoctoral researcher at the Institute for Quantitative Health Sciences & Engineering, Michigan State University, East Lansing, MI
Early infancy is vital for shaping the gut microbiome, which affects neurodevelopment, particularly dendritic complexity. However, it’s unclear if these links are causal. To investigate, we transplanted fecal matter from human infants to germ-free (GF) pregnant mice and studied dendritic morphology changes in the offspring’s brain regions. Pregnant mice were divided into four groups: those given fecal slurry from infants with high Bifidobacterium (HUM1), high Bacteroides (HUM2) levels, from specific pathogen-free mice (SPF) and GF received autoclaved fecal slurry from SPF mice. Gut microbiome groups differed significantly dendritic length in PFC, HUM1 and SPF mice having longer dendrites than GF and HUM2 mice. HUM1 and SPF mice have more dendritic volume than HUM2 mice in Nucleus accumbens (NAc). HUM2 mice have reduced spine density than other groups in hippocampus while in the NAc, HUM2 mice have reduced density compared to HUM1 and SPF mice, and HUM1 mice have greater density than GF mice. In NAc, spiny neurons had higher stubby spine density in HUM1 and SPF animals than GF and HUM2. HUM1 mice had higher mushroom spine density than GF; SPF mice had higher mushroom density than other groups. Similarly, HUM1 had higher filopodia spine density than GF and HUM2 mice in NAc. In hippocampus, pyramidal neurons had longer long thin spine than HUM1 and SPF; and HUM1 and HUM2 had longer thin spine lengths than SPF. This study presents novel insights into how gut microbiomes affect dendritic morphology, suggesting potential benefits for neurodevelopmental outcomes.