Infant guts are alive with rich viral diversity

Having been regarded primarily as pathogens, the metagenomic era has turned our view of bacteria on its head. Bacteria in nature are more diverse than we ever thought, and a diverse gut microbiome protects us from a range of chronic diseases. But what about viruses? Could it be the same for them?
Infant guts are alive with rich viral diversity

CrAssphage is a widespread bacterial virus (or phage) that infects Bacteroides. It was discovered in 2014, almost by accident, because its genome kept popping up in human gut metagenomic data from all over the world. Since then it has been found to be part of a huge viral family and dubbed "the most abundant virus in humans" (PMID 29133882).

At the Copenhagen Prospective Studies for Asthma in Childhood (COPSAC) we study children. Specifically we are trying to decipher how having certain gut bacteria in infancy can protect you from developing chronic disease later in life. In 2016 Prof. Dennis S. Nielsen from Copenhagen University suggested that COPSAC should look into gut viruses as well. After obtaining funding from the EU, Dr. Ling Deng, the joint first-author, spent two years extracting and sequencing viromes from the diapers of 650 1-year-olds.

Human gut viral diversity was largely unexplored at the time, so when I mapped our sequencing reads to online viral databases, almost nothing came up. CrAssphage was there but only in small amounts. Which viruses did the remaining reads come from? After assembling the reads into contigs and matching them to CRISPR spacers in metagenomes from the same children, I spent three years at COPSAC, manually teasing out all of our phage sequences.

In total we found 10,000 viral species, a number that was ten times greater than the number of bacterial species in the children. CrAssphage was also there, but it was overshadowed by other more widespread and abundant viral taxa. By using the ICTV's guidelines for viral taxonomy, we found that these 10,000 viral species could be grouped into almost 250 viral families. Sixteen of those viral families were already known, including human anelloviruses, adenoviruses, crAssphage and Gubaphage. But the vast majority were new and we named them after the children from the COPSAC2010 cohort that delivered the diapers.

Fecal viromes from 647 infants at age 1 year weredeeply sequenced, assembled and curated, resultingin the identification of 10,021 viral species fallingwithin 248 viral families. Click family names to expandfamilies. Keep browser tabs open to save load times.Novel families were named after the infants. Knownfamilies are highlighted in red. Host ranges for eachfamily are given and the families have been groupedinto 17 order-level clusters
Figure 1: An atlas of infant gut DNA virus diversity. Fecal viromes from 647 infants at age 1 year weredeeply sequenced, assembled and curated, resulting in the identification of 10,021 viral species falling within 248 viral families. Host ranges for eachfamily are given and the families have been grouped into 17 virus order-level clusters.

We have made an online version of the paper's Figure 1 ( so anybody can browse the families and download the data. 250 viral families is a lot, and right now we are wondering whether infant guts are particularly rich in viral diversity, or whether adults sport similar numbers, only nobody has done the counting systematically.

One thing we found was that most of the phage families in the healthy infant gut are temperate, meaning they integrate as prophages on the chromosomes of their host bacteria. In contrast, in healthy adults most gut phages are virulent, including crAssphage, Gubaphage, etc. Why is there this difference in viral lifestyles? And could be linked to the apparent difference in viral diversity between infants and adults?

Our current working hypothesis is that viral loads are elevated in the infant gut because the immune system is still maturing. Human viruses, like Anelloviruses are already elevated because the immune system has not yet learned how to target them. Additionally, we know that most gut bacterial species carry at least a few prophages on their genomes, and that bacteria induce their prophages when stressed. The bacteria in the infant gut may be particularly stressed because of their new environment, or because the maturing immune system has still not decided whether or not to leave them alone. Considering that the infants enrolled in COPSAC2010 were healthy, having a gut teeming with viruses must be a natural part of being a baby.

The 230 new viral families were found in faecal samples from 1-year-olds. Next we'll look at the 1-month and 1-week samples from the same children to see whether the high diversity is there already then. Also next up, is our statistical comparison of the viral families to clinical diagnoses for the children. These analyses will reveal whether some viral families are able to protect the children from acquiring chronic diseases such as asthma or ADHD. The advantage of doing such analyses at the viral family-level is that most children carry most families at varying abundances which we can compare to their diagnoses. A species-level analysis, on the other hand, would have been too sparse with individual species being way too rare to yield statistically powerful testing.

Now the children are 12 years old. As they grow older we will acquire more clinical data and be able to test the early life gut virome against more diseases. At COPSAC we do believe that many chronic diseases begin in early childhood even if symptoms appear much later. Our setup allows us to test this. For now at least we know that having a gut teeming with viruses must be a natural part of being a baby.

Please sign in or register for FREE

If you are a registered user on Nature Portfolio Microbiology Community, please sign in