Early life dynamics of the human gut virome and bacterial microbiome in infants - PubMed (original) (raw)

. 2015 Oct;21(10):1228-34.

doi: 10.1038/nm.3950. Epub 2015 Sep 14.

Affiliations

Early life dynamics of the human gut virome and bacterial microbiome in infants

Efrem S Lim et al. Nat Med. 2015 Oct.

Abstract

The early years of life are important for immune development and influence health in adulthood. Although it has been established that the gut bacterial microbiome is rapidly acquired after birth, less is known about the viral microbiome (or 'virome'), consisting of bacteriophages and eukaryotic RNA and DNA viruses, during the first years of life. Here, we characterized the gut virome and bacterial microbiome in a longitudinal cohort of healthy infant twins. The virome and bacterial microbiome were more similar between co-twins than between unrelated infants. From birth to 2 years of age, the eukaryotic virome and the bacterial microbiome expanded, but this was accompanied by a contraction of and shift in the bacteriophage virome composition. The bacteriophage-bacteria relationship begins from birth with a high predator-low prey dynamic, consistent with the Lotka-Volterra prey model. Thus, in contrast to the stable microbiome observed in adults, the infant microbiome is highly dynamic and associated with early life changes in the composition of bacteria, viruses and bacteriophages with age.

PubMed Disclaimer

Conflict of interest statement

Competing financial interests:

The authors declare no competing financial interests.

Figures

Figure 1

Figure 1

Study design and metagenomic analysis of the infant gut virome. (a) Sequencing strategy to characterize the microbiome of 8 healthy infants (4 twin pairs). (b) Heatmap of reads assigned to virus families show that the profile is influenced by the sequencing method. Comparison of representative specimens is shown: fecal specimen from infant A2 at 24 months (A2-24), infant B1 at 6 months (B1-6) and 18 months (B1-18), infant B2 at 6 months (B2-6) and 18 months (B2-18), infant C1 at 6 months (C1-6) and infant C2 at 6 months (C2-6). SIA, sequence independent DNA and RNA amplification; MDA, multiple displacement amplification. (c) Presence-absence heatmap shows the viruses identified by subject (infants A1, A2, B1, B2, C1, C2, D1 and D2) and time point (0, 3, 6, 12, 18 and 24 months) during the first two years of life.

Figure 2

Figure 2

Analysis of virome beta-diversity. Agglomerative hierarchical clustering of Bray-Curtis dissimilarity of virome communities (eukaryotic viruses and bacteriophages; genera) at indicated ages.

Figure 3

Figure 3

Alterations in the eukaryotic RNA and DNA viruses with age and evidence of shared viromes between co-twins. (a) Richness (number of observed taxa) of eukaryotic DNA and RNA virus genera (n = 8 infants). Linear regression, R2 value and 95% confidence intervals are shown. (b) Number of specimens harboring indicated eukaryotic RNA virus genera. (c) Maximum likelihood phylogeny of parechovirus sequences. Bootstrap values are indicated on the branches. (d) Number of specimens harboring indicated eukaryotic DNA virus genera. (e) Phylogenetic relationships of 61 anellovirus contigs and 12 reference strains inferred from the ORF1 amino acid alignment, generated by the maximum likelihood method. Genera are highlighted in indicated colors. (f) Presence-absence heatmap sequencing reads mapped to the anellovirus contigs. Contigs are colored by their genera phylogenetic assignment from (e). (g) Richness of anelloviruses species at indicated age (n = 8 infants) is shown. Statistical significance was assessed by Wilcoxon test (paired, non-parametric); *P = 0.01–0.05, ** P <0.01. (h) Comparison of the proportion of shared anellovirus taxa (genome contigs) acquired during the first two years of life between co-twins (n = 4 co-twin comparisons) and unrelated infants (n = 24 comparison between unrelated infants). Statistical significance was assessed by Student’s t-test; *P < 0.05.

Figure 4

Figure 4

Decrease in bacteriophage richness and diversity with age coincides with a shift in bacteriophage composition. (a) Richness of bacteriophage species is shown (n = 8 infants). Linear regression, R2 value and 95% confidence intervals are shown. (b) Rarefaction curves show the acquisition of bacteriophage species richness (500 permutations). Curves from samples at the same age are indicated in colors. (c) Alpha diversity (Shannon index) of bacteriophage species is plotted (n = 8 infants). Linear regression, R2 value and 95% confidence intervals. (d) Bray-Curtis distance of the bacteriophage virome at the genus level within twin pairs (white) (n = 4 co-twin comparisons) compared to unrelated infants (shaded) (n = 24 comparison between unrelated infants). Statistical significance was assessed by Student’s t-test; *P = 0.01–0.05, **P < 0.01. (e) Relative abundance of bacteriophage families. (f) Plot shows the relationship of the Microviridae family abundance compared to Caudovirales order abundance (n = 48 sampling time points). Linear regression and 95% confidence intervals are shown, and the Spearman correlation coefficient is indicated.

Figure 5

Figure 5

Bacterial community expansion with age. (a) Richness (number of observed bacterial OTUs) of bacterial OTUs(n = 8 infants). Linear regression, R2 value and 95% confidence intervals are shown. (b) Bacterial alpha diversity (Faith’s phylogenetic diversity)(n = 8 infants). Linear regression, R2 value and 95% confidence intervals are shown. (c) Relative abundance of bacterial families based on 16S rRNA gene sequences. (d) Unifrac distance of the bacterial community compared within twin pairs (white) (n = 4 co-twin comparisons) and between unrelated infants (shaded) (n = 24 comparison between unrelated infants). Statistical significance was assessed by Student’s t-test; **P < 0.01.

Figure 6

Figure 6

Inverse relationships between bacteriophage and bacteria. (a) Correlation between bacteriophage diversity and bacterial diversity (n = 48 sampling time points). Line indicates linear regression, and the Spearman correlation coefficient is shown. Color spectrum indicates age progression from 0–24 months. (b) Correlation plot between bacteriophage richness and bacterial richness(n = 48 sampling time points). Line indicates linear regression, Spearman correlation coefficient is shown. Color spectrum indicates age progression from 0–24 months.

Comment in

References

    1. Virgin HW. The virome in mammalian physiology and disease. Cell. 2014;157:142–150. - PMC - PubMed
    1. Norman JM, Handley SA, Virgin HW. Kingdom-agnostic metagenomics and the importance of complete characterization of enteric microbial communities. Gastroenterology. 2014;146:1459–1469. - PMC - PubMed
    1. Qin N, et al. Alterations of the human gut microbiome in liver cirrhosis. Nature. 2014;513:59–64. - PubMed
    1. Cho I, Blaser MJ. The human microbiome: at the interface of health and disease. Nat Rev Genet. 2012;13:260–270. - PMC - PubMed
    1. Littman DR, Pamer EG. Role of the commensal microbiota in normal and pathogenic host immune responses. Cell Host Microbe. 2011;10:311–323. - PMC - PubMed

Publication types

MeSH terms

Grants and funding

LinkOut - more resources