Temporal development of the gut microbiome in early childhood from the TEDDY study

doi:10.1038/s41586-018-0617-x

. 2018 Oct;562(7728):583-588.

doi: 10.1038/s41586-018-0617-x. Epub 2018 Oct 24.

Temporal development of the gut microbiome in early childhood from the TEDDY study

Christopher J Stewart^{1

2}, Nadim J Ajami³, Jacqueline L O'Brien³, Diane S Hutchinson³, Daniel P Smith³, Matthew C Wong³, Matthew C Ross³, Richard E Lloyd³, HarshaVardhan Doddapaneni⁴, Ginger A Metcalf⁴, Donna Muzny⁴, Richard A Gibbs⁴, Tommi Vatanen⁵, Curtis Huttenhower⁵, Ramnik J Xavier⁵, Marian Rewers⁶, William Hagopian⁷, Jorma Toppari^{8

9}, Anette-G Ziegler^{10

11

12}, Jin-Xiong She¹³, Beena Akolkar¹⁴, Ake Lernmark¹⁵, Heikki Hyoty^{16

17}, Kendra Vehik¹⁸, Jeffrey P Krischer¹⁸, Joseph F Petrosino¹⁹

Affiliations

¹ Alkek Center for Metagenomics and Microbiome Research, Department of Molecular Virology and Microbiology, Baylor College of Medicine, Houston, TX, USA. christopher.stewart@ncl.ac.uk.
² Institute of Cellular Medicine, Newcastle University, Newcastle upon Tyne, UK. christopher.stewart@ncl.ac.uk.
³ Alkek Center for Metagenomics and Microbiome Research, Department of Molecular Virology and Microbiology, Baylor College of Medicine, Houston, TX, USA.
⁴ Human Genome Sequencing Center, Baylor College of Medicine, Houston, TX, USA.
⁵ Broad Institute of MIT and Harvard, Cambridge, MA, USA.
⁶ Barbara Davis Center for Childhood Diabetes, University of Colorado, Aurora, CO, USA.
⁷ Pacific Northwest Research Institute, Seattle, WA, USA.
⁸ Institute of Biomedicine, Research Centre for Integrative Physiology and Pharmacology, University of Turku, Turku, Finland.
⁹ Department of Pediatrics, Turku University Hospital, Turku, Finland.
¹⁰ Institute of Diabetes Research, Helmholtz Zentrum München, Munich, Germany.
¹¹ Forschergruppe Diabetes, Technische Universität München, Klinikum Rechts der Isar, Munich, Germany.
¹² Forschergruppe Diabetes e.V. at Helmholtz Zentrum München, Munich, Germany.
¹³ Center for Biotechnology and Genomic Medicine, Medical College of Georgia, Augusta University, Augusta, GA, USA.
¹⁴ National Institute of Diabetes & Digestive & Kidney Diseases, Bethesda, MD, USA.
¹⁵ Department of Clinical Sciences, Lund University/CRC, Skane University Hospital, Malmö, Sweden.
¹⁶ Department of Virology, Faculty of Medicine and Biosciences, University of Tampere, Tampere, Finland.
¹⁷ Fimlab Laboratories, Pirkanmaa Hospital District, Tampere, Finland.
¹⁸ Health Informatics Institute, Morsani College of Medicine, University of South Florida, Tampa, FL, USA.
¹⁹ Alkek Center for Metagenomics and Microbiome Research, Department of Molecular Virology and Microbiology, Baylor College of Medicine, Houston, TX, USA. jpetrosi@bcm.edu.

PMID: 30356187
PMCID: PMC6415775
DOI: 10.1038/s41586-018-0617-x

Temporal development of the gut microbiome in early childhood from the TEDDY study

Christopher J Stewart et al. Nature. 2018 Oct.

. 2018 Oct;562(7728):583-588.

doi: 10.1038/s41586-018-0617-x. Epub 2018 Oct 24.

Authors

Affiliations

¹ Alkek Center for Metagenomics and Microbiome Research, Department of Molecular Virology and Microbiology, Baylor College of Medicine, Houston, TX, USA. christopher.stewart@ncl.ac.uk.
² Institute of Cellular Medicine, Newcastle University, Newcastle upon Tyne, UK. christopher.stewart@ncl.ac.uk.
³ Alkek Center for Metagenomics and Microbiome Research, Department of Molecular Virology and Microbiology, Baylor College of Medicine, Houston, TX, USA.
⁴ Human Genome Sequencing Center, Baylor College of Medicine, Houston, TX, USA.
⁵ Broad Institute of MIT and Harvard, Cambridge, MA, USA.
⁶ Barbara Davis Center for Childhood Diabetes, University of Colorado, Aurora, CO, USA.
⁷ Pacific Northwest Research Institute, Seattle, WA, USA.
⁸ Institute of Biomedicine, Research Centre for Integrative Physiology and Pharmacology, University of Turku, Turku, Finland.
⁹ Department of Pediatrics, Turku University Hospital, Turku, Finland.
¹⁰ Institute of Diabetes Research, Helmholtz Zentrum München, Munich, Germany.
¹¹ Forschergruppe Diabetes, Technische Universität München, Klinikum Rechts der Isar, Munich, Germany.
¹² Forschergruppe Diabetes e.V. at Helmholtz Zentrum München, Munich, Germany.
¹³ Center for Biotechnology and Genomic Medicine, Medical College of Georgia, Augusta University, Augusta, GA, USA.
¹⁴ National Institute of Diabetes & Digestive & Kidney Diseases, Bethesda, MD, USA.
¹⁵ Department of Clinical Sciences, Lund University/CRC, Skane University Hospital, Malmö, Sweden.
¹⁶ Department of Virology, Faculty of Medicine and Biosciences, University of Tampere, Tampere, Finland.
¹⁷ Fimlab Laboratories, Pirkanmaa Hospital District, Tampere, Finland.
¹⁸ Health Informatics Institute, Morsani College of Medicine, University of South Florida, Tampa, FL, USA.
¹⁹ Alkek Center for Metagenomics and Microbiome Research, Department of Molecular Virology and Microbiology, Baylor College of Medicine, Houston, TX, USA. jpetrosi@bcm.edu.

PMID: 30356187
PMCID: PMC6415775
DOI: 10.1038/s41586-018-0617-x

Abstract

The development of the microbiome from infancy to childhood is dependent on a range of factors, with microbial-immune crosstalk during this time thought to be involved in the pathobiology of later life diseases^1-9 such as persistent islet autoimmunity and type 1 diabetes^10-12. However, to our knowledge, no studies have performed extensive characterization of the microbiome in early life in a large, multi-centre population. Here we analyse longitudinal stool samples from 903 children between 3 and 46 months of age by 16S rRNA gene sequencing (n = 12,005) and metagenomic sequencing (n = 10,867), as part of the The Environmental Determinants of Diabetes in the Young (TEDDY) study. We show that the developing gut microbiome undergoes three distinct phases of microbiome progression: a developmental phase (months 3-14), a transitional phase (months 15-30), and a stable phase (months 31-46). Receipt of breast milk, either exclusive or partial, was the most significant factor associated with the microbiome structure. Breastfeeding was associated with higher levels of Bifidobacterium species (B. breve and B. bifidum), and the cessation of breast milk resulted in faster maturation of the gut microbiome, as marked by the phylum Firmicutes. Birth mode was also significantly associated with the microbiome during the developmental phase, driven by higher levels of Bacteroides species (particularly B. fragilis) in infants delivered vaginally. Bacteroides was also associated with increased gut diversity and faster maturation, regardless of the birth mode. Environmental factors including geographical location and household exposures (such as siblings and furry pets) also represented important covariates. A nested case-control analysis revealed subtle associations between microbial taxonomy and the development of islet autoimmunity or type 1 diabetes. These data determine the structural and functional assembly of the microbiome in early life and provide a foundation for targeted mechanistic investigation into the consequences of microbial-immune crosstalk for long-term health.

PubMed Disclaimer

Conflict of interest statement

The authors declare no competing interests.

Figures

**Fig. 1. DMM clustering of 16S rRNA gene sequencing data (n = 12,005).**
The entire dataset formed ten distinct clusters based on lowest Laplace approximation. a, Heat map showing the relative abundance of the 25 most dominant bacterial genera per DMM cluster. Taxa names in square brackets are in need of formal taxonomic revision. b, Box plots showing the alpha diversity (richness and Shannon’s diversity) per each DMM cluster. The centre line denotes the median, the boxes cover the 25th and 75th percentiles, and the whiskers extend to the most extreme data point, which is no more than 1.5 times the length of the box away from the box. Points outside the whiskers represent outlier samples. c, Transition model showing the progression of samples through each DMM cluster per each time point, from months 3 to 46 of life. Dashed boxes show the three phases of microbiome progression (developmental, transitional and stable phase). Solid squares next to the labels denote the significant changes in phyla and Shannon’s diversity (H') per phase based on multiple linear regression. All phyla and the H' were significant in the developmental phase, two phyla and the H' were significant in the transitional phase, and no phyla or the H' were in the stable phase. Nodes and edges are sized based on the total counts. Nodes are coloured according to DMM cluster number and edges are coloured by the transition frequency. Transitions with less than 4% frequency are not shown. Results are further supported by the metagenomic sequencing data in Extended Data Fig. 2.

**Fig. 2. Significance and explained variance of 22 microbiome covariates modelled by EnvFit across all data types.**
Horizontal bars show the amount of variance (r²) explained by each covariate in the model as determined by EnvFit. The groups within each covariate are detailed in Extended Data Table 1. Covariates are coloured based on overall metadata group. Significant covariates (false discovery rate (FDR) P < 0.05) are represented in bold font. Asterisk denotes the significant covariates at each time point. BMI, body mass index; wtgain, weight gain. a, Microbiome profiles at the genus level based on 16S rRNA gene sequencing data (n = 4,069). b, Microbiome profiles at the species level based on metagenomic sequencing (n = 3,843). c, Functional metagenomic capacity at the module level based on metagenomic sequencing (n = 3,843).

**Fig. 3. Breastfeeding status was the most significant microbiome covariate associated with all datasets throughout the first year of life.**
Breastfeeding status was significantly associated with microbiome profiles over the first three time points (months 3–14, n = 2,257; Supplementary Table 1). Curves show locally weighted scatterplot smoothing (LOESS) for the data per category, and shaded areas show permutation-based 95% confidence intervals for the fit. a, Non-metric multidimensional scaling (NMDS) ordination plots showing the mean centroid of each breastfeeding status group. Plots include only the first sample obtained from a patient within a given time point; months 3–6, 7–10 and 11–14. Centroid size based on number of samples and the bars represent the ±95% confidence interval. b, Plots showing the receipt of breast milk from months 3 to 40 of age compared to the relative abundance of the six most abundant *Bifidobacterium* species over the same period (n = 11,717). c, Longitudinal Shannon diversity index from months 3 to 40 of age (n = 11,717). d, Longitudinal development of the microbiome maturation based on the microbiota age and MAZ score against the age of the infant at sampling (n = 11,717). e, Heat map showing the mean abundance of all significant modules as determined by MaAsLin analysis at each of the first three time points. The corresponding pathway for each module is also presented. BM, breast milk. f, Stacked bar plots showing the abundance of each significant module binned at the pathway level. Abundance plotted per bacterial species, with the five most significant species associated with breastfed and non-breastfed infants, respectively.

**Extended Data Fig. 1. Characterization of the gut microbiome over the first 40 months of life (n = 11,717).**
a–d, 16S rRNA gene sequencing (a–c) and metagenomic sequencing (d) analysis. Curves show LOESS fit for the data per category, and shaded areas show permutation-based 95% confidence intervals for the fit. a, Summary of overall dietary status. b, The mean alpha diversity (richness and Shannon diversity) per child increased rapidly from 3 to 20 months of life. c, The mean relative abundance of the five most abundant bacterial phyla show changes from 3 to 20 months of life and generally remain stable after month 30 of life. d, The mean relative abundance of the ten most abundant bacterial pathways shows relative stability, with ABC transporters and two-component system showing the largest reduction from 3 to 20 months of life.

**Extended Data Fig. 2. DMM clustering of metagenomic sequencing data (n = 10,867).**
The entire dataset formed 18 distinct clusters based on lowest Laplace approximation. a, Heat map showing the relative abundance of the 25 most dominant bacterial species per each DMM cluster. b, Box plots showing the alpha diversity (richness and Shannon’s diversity) for each DMM cluster. The centre line shows the median, the boxes cover the 25th and 75th percentiles, and the whiskers extend to the most extreme data point, which is no more than 1.5 times the length of the box away from the box. Points outside the whiskers represent outlier samples. c, Transition model showing the progression of samples through each DMM cluster per each time point, from months 3 to 46 of life. Dashed boxes show the three phases of microbiome progression (developmental, transitional and stable phase). Solid squares next to the labels denote the significant changes in phyla and Shannon diversity (H') per phase based on multiple linear regression. All phyla and the H' were significant in the developmental phase, two phyla and the H' were significant in the transitional phase, and no phyla or the H' were in the stable phase. Nodes and edges are sized based on the total counts. Nodes are coloured according to DMM cluster number and edges are coloured by the transition frequency. Transitions with less than 2% frequency were omitted from the plot.

**Extended Data Fig. 3. Twenty bacterial OTUs classified by random forest regression analysis as most age discriminatory over the first 40 months of life.**
Rank importance of OTUs determined by applying the random forest regression to the chronological age of 150 full-term, vaginally delivered, breastfed infants (n = 2,871 stool samples). The importance of OTUs is determined by the percentage increase in mean-squared error of microbiota age prediction when the relative abundance of each OTU were randomly permuted (mean importance ± s.d., n = 100 replicates). These selected OTUs explained 72% of the variance (compared to 75% variance explained with all OTUs in model) and were used to define maturation of the gut microbiome by microbiota age and MAZ score. OTUs are named to the genus level and coloured based on association with life stage; blue were associated with samples collected in the first 15 months, green with samples collected between months 15 and 30, and red were with samples collected after month 30. a, Twenty OTUs ranked by importance to the accuracy of the model. The tenfold cross-validation error is also displayed in order of variable importance. Blue dotted line represents the 20 OTUs used in the model. b, Heat map of mean relative abundance of the 20 selected OTUs per month from 3 to 40 months of age.

**Extended Data Fig. 4. The microbiota was not associated with the development of persistent IA and T1D.**
Data are based on 16S rRNA gene sequencing (n = 11,717). Analysis based on a nested 1:1 case–control cohort of equal samples. Curves show LOESS fit for the data per category, and shaded areas show permutation-based 95% confidence intervals for the fit. a, b, The number of OTUs (a) and the Shannon’s diversity index (b) in the IA cohort. c, d, The number of OTUs (c) and Shannon’s diversity (d) in the T1D cohort. e, f, Microbiota age (e) and MAZ score (f) in the IA cohort. g, h, Microbiota age (g) and MAZ score (h) in the T1D cohort. i, j, Forest plot showing the odds ratios for the association between the microbiome stability metrics and development of IA (i) and T1D (j). A separate conditional logistic regression was run for four time intervals: (1) birth to onset; (2) 12 months before onset; (3) 6–12 months before onset; and (4) 6 months before onset. Models were adjusted for HLA genotype, mode of delivery, duration of exclusive breastfeeding, number of antibiotic courses, and number of infectious episodes. Community states are the total number of unique clusters exhibited by an infant and state transitions are the number of transitions between clusters. No odds ratio was significantly different between cases and controls (Supplementary Table 3).

**Extended Data Fig. 5. Association of the gut microbiome with birth mode.**
Birth mode was significantly associated with the microbiome in months 3–6 by 16S rRNA gene sequencing and in all time points up to month 14 by metagenomic sequencing (see Supplementary Table 1). Curves show LOESS fit for the data per category, and shaded areas show permutation-based 95% confidence intervals for the fit. a, Longitudinal development of the *Bacteroides* genus as determined by 16S rRNA gene sequencing (n = 11,717). b, Longitudinal development of the six most abundant species within the *Bacteroides* genera as determined by metagenomic sequencing (n = 10,867). Grid overlay added to aid visual interpretation. c, NMDS ordination plots showing the mean centroid of each birth mode group stratified by *Bacteroides* positive or negative based on detection 16S rRNA gene sequencing. Plots include only the first sample obtained from a patient within a given time point for months 3–6, 7–10 and 11–14 (n = 2,257). Centroid size based on number of samples and the bars represent the ±95% confidence interval. d, Longitudinal development of the alpha diversity (richness and Shannon’s diversity) with birth mode further stratified according to *Bacteroides* positive or negative (n = 11,717). e, Longitudinal development of the microbiome maturation based on the microbiota age and MAZ score against the age of the infant at sampling (n = 11,717). Birth mode was further stratified according to *Bacteroides* positive or negative.

**Extended Data Fig. 6. The relative abundance of *Bacteroides* stratified by breast milk and geographical location.**
Curves show LOESS fit for the data per category, and shaded areas show permutation-based 95% confidence intervals for the fit. a, b, *Bacteroides* genera based on 16S rRNA gene sequencing data (n = 11,717) stratified by breast milk status (a) and geographical location (b). c, The top 6 *Bacteroides* species based on metagenomic sequencing data (n = 10,867) stratified by geographical location.

**Extended Data Fig. 7. Environmental covariates significantly associated with the microbiome profiles.**
16S rRNA gene sequencing data plotted from months 3–40 of life (n = 11,717). Curves show LOESS fit for the data per category, and shaded areas show permutation-based 95% confidence intervals for the fit. Significance determined by linear mixed-effects models in accordance with observed phases of maturation: developmental (months 3–14), transitional (months 15–30), and stable (months 31–46). Shaded lines represent the ±95% confidence interval. Longitudinal development of the Shannon’s diversity index, microbiota age and MAZ score by geographical location (a–c), occurrence of household siblings (d–f), and occurrence of household furry pets (g–i).

See this image and copyright information in PMC

Cited by

Transcriptional networks in at-risk individuals identify signatures of type 1 diabetes progression.
Xhonneux LP, Knight O, Lernmark Å, Bonifacio E, Hagopian WA, Rewers MJ, She JX, Toppari J, Parikh H, Smith KGC, Ziegler AG, Akolkar B, Krischer JP, McKinney EF. Xhonneux LP, et al. Sci Transl Med. 2021 Mar 31;13(587):eabd5666. doi: 10.1126/scitranslmed.abd5666. Sci Transl Med. 2021. PMID: 33790023 Free PMC article.
Early Onset of Autoimmune Diabetes in Children with Down Syndrome-Two Separate Aetiologies or an Immune System Pre-Programmed for Autoimmunity?
Mortimer GL, Gillespie KM. Mortimer GL, et al. Curr Diab Rep. 2020 Aug 25;20(9):47. doi: 10.1007/s11892-020-01318-8. Curr Diab Rep. 2020. PMID: 32839884 Free PMC article. Review.
Milk microbiome transplantation: recolonizing donor milk with mother's own milk microbiota.
Stinson LF, Ma J, Lai CT, Rea A, Perrella SL, Geddes DT. Stinson LF, et al. Appl Microbiol Biotechnol. 2024 Dec;108(1):74. doi: 10.1007/s00253-023-12965-8. Epub 2024 Jan 9. Appl Microbiol Biotechnol. 2024. PMID: 38194146 Free PMC article.
Food-breastmilk combinations alter the colonic microbiome of weaning infants: an in silico study.
da Silva VG, Smith NW, Mullaney JA, Wall C, Roy NC, McNabb WC. da Silva VG, et al. mSystems. 2024 Sep 17;9(9):e0057724. doi: 10.1128/msystems.00577-24. Epub 2024 Aug 27. mSystems. 2024. PMID: 39191378 Free PMC article.
The right educational environment: Oral tolerance in early life.
Cheifetz TR, Knoop KA. Cheifetz TR, et al. Immunol Rev. 2024 Sep;326(1):17-34. doi: 10.1111/imr.13366. Epub 2024 Jul 13. Immunol Rev. 2024. PMID: 39001685 Review.

See all "Cited by" articles

References

1. Yuan C, et al. Association between cesarean birth and risk of obesity in offspring in childhood, adolescence, and early adulthood. JAMA Pediatr. 2016;170:e162385. doi: 10.1001/jamapediatrics.2016.2385. - DOI - PMC - PubMed
1. Sevelsted A, Stokholm J, Bønnelykke K, Bisgaard H. Cesarean section and chronic immune disorders. Pediatrics. 2015;135:e92–e98. doi: 10.1542/peds.2014-0596. - DOI - PubMed
1. Mayer-Davis EJ, et al. Breast-feeding and risk for childhood obesity: does maternal diabetes or obesity status matter? Diabetes Care. 2006;29:2231–2237. doi: 10.2337/dc06-0974. - DOI - PMC - PubMed
1. Klement E, Cohen RV, Boxman J, Joseph A, Reif S. Breastfeeding and risk of inflammatory bowel disease: a systematic review with meta-analysis. Am. J. Clin. Nutr. 2004;80:1342–1352. doi: 10.1093/ajcn/80.5.1342. - DOI - PubMed
1. Koplin JJ, et al. Environmental and demographic risk factors for egg allergy in a population-based study of infants. Allergy. 2012;67:1415–1422. doi: 10.1111/all.12015. - DOI - PubMed

Publication types

Actions
Actions
Actions

MeSH terms

Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions

Substances

Actions

Grants and funding

LinkOut - more resources

Full Text Sources
Other Literature Sources
- H1 Connect
- The Lens - Patent Citations Database

[1] Yuan C, et al. Association between cesarean birth and risk of obesity in offspring in childhood, adolescence, and early adulthood. JAMA Pediatr. 2016;170:e162385. doi: 10.1001/jamapediatrics.2016.2385. - DOI - PMC - PubMed

[2] Yuan C, et al. Association between cesarean birth and risk of obesity in offspring in childhood, adolescence, and early adulthood. JAMA Pediatr. 2016;170:e162385. doi: 10.1001/jamapediatrics.2016.2385. - DOI - PMC - PubMed

[3] Sevelsted A, Stokholm J, Bønnelykke K, Bisgaard H. Cesarean section and chronic immune disorders. Pediatrics. 2015;135:e92–e98. doi: 10.1542/peds.2014-0596. - DOI - PubMed

[4] Sevelsted A, Stokholm J, Bønnelykke K, Bisgaard H. Cesarean section and chronic immune disorders. Pediatrics. 2015;135:e92–e98. doi: 10.1542/peds.2014-0596. - DOI - PubMed

[5] Mayer-Davis EJ, et al. Breast-feeding and risk for childhood obesity: does maternal diabetes or obesity status matter? Diabetes Care. 2006;29:2231–2237. doi: 10.2337/dc06-0974. - DOI - PMC - PubMed

[6] Mayer-Davis EJ, et al. Breast-feeding and risk for childhood obesity: does maternal diabetes or obesity status matter? Diabetes Care. 2006;29:2231–2237. doi: 10.2337/dc06-0974. - DOI - PMC - PubMed

[7] Klement E, Cohen RV, Boxman J, Joseph A, Reif S. Breastfeeding and risk of inflammatory bowel disease: a systematic review with meta-analysis. Am. J. Clin. Nutr. 2004;80:1342–1352. doi: 10.1093/ajcn/80.5.1342. - DOI - PubMed

[8] Klement E, Cohen RV, Boxman J, Joseph A, Reif S. Breastfeeding and risk of inflammatory bowel disease: a systematic review with meta-analysis. Am. J. Clin. Nutr. 2004;80:1342–1352. doi: 10.1093/ajcn/80.5.1342. - DOI - PubMed

[9] Koplin JJ, et al. Environmental and demographic risk factors for egg allergy in a population-based study of infants. Allergy. 2012;67:1415–1422. doi: 10.1111/all.12015. - DOI - PubMed

[10] Koplin JJ, et al. Environmental and demographic risk factors for egg allergy in a population-based study of infants. Allergy. 2012;67:1415–1422. doi: 10.1111/all.12015. - DOI - PubMed

Save citation to file

Email citation

Add to Collections

Add to My Bibliography

Your saved search

Create a file for external citation management software

Your RSS Feed

Temporal development of the gut microbiome in early childhood from the TEDDY study

Affiliations

Temporal development of the gut microbiome in early childhood from the TEDDY study

Authors

Affiliations

Abstract

Conflict of interest statement

Figures

Similar articles

Cited by

References

Publication types

MeSH terms

Substances

Grants and funding

LinkOut - more resources

Full Text Sources

Other Literature Sources

Abstract

Conflict of interest statement

Figures

Similar articles

Cited by

References

Publication types

MeSH terms

Substances

Related information

Grants and funding

LinkOut - more resources

Full Text Sources

Other Literature Sources