A changing gut virome ecological landscape with longevity

Abstract

A recent study shows how the ecological composition of the human gut virome changes with age, showing a decline in core taxa and an enrichment of sub-dominant taxa, similar to what has been observed in the gut bacteriomes of centenarians.

The human gut virome remains a largely unexplored, but potentially physiologically important, component of human health. In their recent study, Johansen et al. [1] re-analyzed stool metagenomics data from 172 centenarians and over 190 younger controls to reveal novel viral signatures associated with aging and longevity. Their study highlights a vast and mostly uncharacterized viral diversity in the human gut, and points to strikingly similar virome aging patterns to those previously observed in gut bacterial communities [2]. Johansen et al. further identify the virome as a significant contributor to changes in the metabolic capacity of the gut observed in long-lived individuals, opening possibilities for a better mechanistic understanding of how the aging gut ecosystem may ultimately contribute to longevity.

In order to quantify prophages, Johansen et al. mined gut bacterial metagenomic assembled genomes (MAGs) for viral DNA. The resulting viral contigs were clustered into 4,422 unique viral operational taxonomic units (vOTUs), a large fraction of which (~40%) could not be mapped to a reference database. Comparison of the resulting vOTUs across centenarians (100+ yrs), older (60–100 yrs) and younger (18–60 yrs) adults revealed a gradient of increasing viral diversity with increasing age. The rise in viral diversity was accompanied by a loss of core vOTUs (commonly shared across individuals), and an increasing dominance of rare (<10% prevalence in the cohort), previously uncharacterized vOTUs (Figure 1). This aging pattern was remarkably similar to previously reported signatures of healthy aging in the gut bacteriome, characterized by a depletion of core commensal gut bacteria (e.g., Bacteroides and Faecalibacterium sp.) and a rise in dominance of other less prevalent taxa (i.e., Akkermansia, Desulfovibrio and Clostridium sp.) [2,3].

Figure 1. A changing gut virome throughout the human lifespan.

Younger adults are characterized by gut ecosystems enriched in core prophages that reside within core bacterial hosts in the gut, like Bacteroides and Faecalibacterium species. In later adulthood, we see a decline in core viral and bacterial clades, and a rise in subdominant taxa. These changes correspond with a decline in host mucus production with age, which may lead to a gradual thinning of the mucus layer. Finally, these ecological changes are coupled with functional development of the gut with age, like the relative rise in amino acid fermentation among bacteria and a rise in hydrogen sulfide production genes among phages.

Indeed, there is a clear correlation between previously observed gut bacterial signatures of aging and the accompanying viral footprint reported here [1]. Analyses pairing viruses with their likely bacterial hosts, via CRISPR spacers and prophage integration events, showed that viruses enriched in centenarians were often associated with bacterial taxa that are commonly retained or enriched in long-lived individuals (e.g., Clostridium sp.), while depleted viruses, relative to younger controls, were often associated with core taxa that appear to decline with age (e.g., Bacteroides and Faecalibacterium sp.). As the authors suggest, this viral signature may simply reflect changes in gut microbiome composition, where lysogenic viral dominance is strongly coupled to the ecological dominance of its host (‘piggybacking the winner’). While the environmental drivers remain unclear, this rise in the dominance of non-core gut bacteria and viruses with aging could be an important eco-evolutionary mechanism for the adaptation of the gut ecosystem to the changing physiological landscape of its host. Whether this phenomenon contributes to longevity, is a consequence of aging, or a combination of both remains unclear.

Johansen et al. inferred the rate of viral lysis of cells from the viral-to-bacterial ratio (VBR), with higher VBRs indicating greater lytic activity. Across adults, they observed higher VBRs, and thus more lytic activity, with increasing age. While VBRs have previously been shown to increase with inflammation in the gut, no significant associations were found between inflammatory biomarkers and VBR estimates. Adding further complexity to the observed pattern was a comparison of VBR estimates calculated across 550 infants from Finland, profiled from birth until one year of age, which found higher VBRs in the first year of life than any other age group, including centenarians. It is not yet clear why lytic activity follows a u-shaped curve across the human lifespan, although it may be driven in part by the increased relative abundance of lysogens (mainly within hosts from the Bacteroidetes and Firmicutes phyla) with the acquisition of an adult-like microbiota in the first few years of life, followed by the subsequent decline of these core taxa in long-lived individuals. Future studies pairing virome data with in depth health assessments and other ‘omics measurements (e.g., metabolomics or proteomics) may shed additional light on the forces that tip the balance between lytic and lysogenic life cycles in the gut.

Viruses may augment gut bacterial metabolism through their encoding of auxiliary metabolic genes (AMGs). A focused analysis of these AMGs in centenarians by the Johansen et al. identified an increased contribution of viral DNA to the total pool of genes encoding enzymes responsible for conversion of methionine to homocysteine, sulfate to sulfide (H2S), and taurine to H2S (Figure 1). These results build upon prior reports of longevity-associated functional shifts in gut microbial metabolism. For example, analysis of the same dataset of Japanese centenarians has previously identified novel bile acid synthesis pathways responsible for the synthesis of lithocholic acid (LCA) derivatives (e.g., isoallo-LCA and 3-oxoalloLCA), which were enriched in long-lived individuals relative to younger controls [4]. A gut metagenomics study of Sardinian centenarians demonstrated a significant increase in aromatic amino acid degradation pathways in long-lived individuals relative to younger controls [5]. Consistently, increased concentrations of microbial tryptophan and phenylalanine degradation products were observed in the plasma and serum of individuals approaching extreme old age [2,6]. The present study provides an exciting new perspective on these previous studies, where the gut virome follows a similar taxonomic pattern across the lifespan, while contributing a distinct functional shift in gut metabolic capacity from what is seen in youth. As the authors mention in their report, ~75% of viral proteins are presently unannotated, highlighting the potential for gaining future insights into how the virome may impact or be impacted by host aging.

The observed virome-mediated functional changes associated with longevity (increased H2S synthesis) have conflicting linkages to human health and disease. Microbial production of H2S has been shown to promote colonization resistance and support gut mucosal integrity, both key components of a healthy microbiome that decline with age [7]. However, excessive H2S production may promote mucosal inflammation, with one of the better known H2S producers, Fusobacterium nucleatum, implicated in a number of inflammatory diseases [8]. The context-dependence of these longevity-associated metabolic changes in the gut lends credence to the idea of “right place, right time” for a healthy gut ecosystem [9]. Having high H2S production in the gut may be beneficial in the presence of increased oxidative stress and a thinning mucus layer, as is the case in older age, but have a less beneficial or detrimental effect in a younger gut ecosystem which resides within a vastly different host physiological environment. Identifying which key features of the gut microbiome and virome observed in long-lived individuals can be harnessed to promote healthy aging in younger adults will be a major challenge and opportunity for the field in the years to come.

While the results of the current study provide a compelling glimpse into longevity-associated changes observed in the gut virome, they also highlight the limitations of available databases when it comes to annotating viral genomes. While this annotation problem poses an obstacle, it also suggests that there are many future discoveries waiting to be made into how the gut virome contributes to host health across the lifespan. It is also important to distinguish aging and aging-related diseases from healthy aging, which was not a focus of this particular study. While centenarians are considered a prime model of healthy aging, living a long life in the absence of major diseases and comorbidities, they are not exempt from age-associated decline at the time of sampling [10]. Indeed, even stratifying centenarians and older adults by measures of physical and functional health can reveal distinct gut ecosystem aging dynamics that are masked in a pooled analysis [2]. Future studies that can integrate gut microbiome and virome data with a wealth of phenotypic information on the host health state are likely to provide an even greater resolution to the gut viral signatures reported in this study.