Archaea Are Rare and Uncommon Members of the Mammalian Skin Microbiome

Abstract

Although previous research demonstrates that skin-associated archaea are rarely detected within human skin microbiome data, exist at relatively low abundance, and are primarily affiliated with the Methanobacteriota and Halobacteriota phyla, other studies suggest that archaea are consistently detected and relatively abundant on human skin, with skin "archaeomes" dominated by putative ammonia oxidizers of the Nitrososphaeria class (Thermoproteota phylum, formerly Thaumarchaeota). Here, we evaluated new and existing 16S rRNA gene sequence data sourced from mammalian skin and skin-associated surfaces and generated with two commonly used universal prokaryotic primer sets to assess archaeal prevalence, relative abundance, and taxonomic distribution. Archaeal 16S rRNA gene sequences were detected in only 17.5% of 1,688 samples by high-throughput sequence data, with most of the archaeon-positive samples associated with nonhuman mammalian skin. Only 5.9% of human-associated skin sample data sets contained sequences affiliated with archaeal 16S rRNA genes. When detected, the relative abundance of sequences affiliated with archaeal amplicon sequence variants (ASVs) was less than 1% for most mammalian skin samples and did not exceed 2% for any samples. Although several computer keyboard microbial profiles were dominated by Nitrososphaeria sequences, all other skin microbiome data sets tested were primarily composed of sequences affiliated with Methanobacteriota and Halobacteriota phyla. Our findings revise downward recent estimates of human skin archaeal distributions and relative abundances, especially those affiliated with the Nitrososphaeria, reflecting a limited and infrequent archaeal presence within the mammalian skin microbiome. IMPORTANCE The current state of research on mammalian skin-associated archaea is limited, with the few papers focusing on potential skin archaeal communities often in disagreement with each other. As such, there is no consensus on the prevalence or taxonomic composition of archaea on mammalian skin. Mammalian skin health is in part influenced by its complex microbiota and consortium of bacteria and potential archaea. Without a clear foundational analysis and characterization of the mammalian skin archaeome, it will be difficult for future research to explore the potential impact of skin-associated archaea on skin health and function. The current work provides a much-needed analysis of the mammalian skin archaeome and contributes to building a foundation from which further discussion and exploration of the skin archaeome might continue.

Keywords: 16S RNA; archaea; built environment; gene sequencing; mammalian skin; microbiome; rare biosphere.

FIG 1

Comparison of the relative abundances (A) and ASVs (B) of archaea on human and nonhuman mammalian skin. Samples were sequenced using primer pairs 515F-Y/926R and Pro341F/Pro805R and their relative abundances compared, with pairs indicated with alternating gray/white background bars. The relative abundances of the ASVs are indicated by both percentage and size of the point. Red stars at the bottom indicate samples with more than five archaeal reads; all unmarked samples had four or fewer reads.

FIG 2

(A) Archaeal 16S rRNA gene relative abundances. The relative abundances of archaeal sequences were calculated by dividing the number of sequences affiliated with archaeal ASVs by the total number of sequences for each sample. Relative abundance averages for all samples in each data set are indicated by orange squares within the boxplot. (B) The taxonomic proportions of the archaeome of the skin and skin-associated surfaces are separated by phylum or class. Archaeal taxonomic proportions include archaeal 16S rRNA gene reads only and represent the proportions of archaeal reads belonging to each phylum or class. Class Nitrososphaeria was separated from the phylum Thermoproteota to highlight putative AOA-associated archaea specifically. The Thermoproteota category thus does not contain any Nitrososphaeria-associated reads.

FIG 3

Distribution of all archaeal genera across skin and skin-associated surfaces. The ASV table was collapsed to the genus level and then filtered for archaeal taxa and contains any sample with a nonzero number of archaeal 16S rRNA gene reads. The sizes of the bubbles represent the relative abundances of the genera with respect to the total number of archaeal 16S rRNA gene reads within a sample. Archaeal ASVs not resolved to the genus level were collapsed to their most resolved taxonomic level and are indicated with asterisks, and they are referred to as “genera” in Results and Discussion. A human skin data set (32) was included for comparison.

FIG 4

The prevalences and overlaps of archaeal ASVs on skin and skin-associated surfaces. The tree contains all archaeal ASVs from each data set. ASVs in black are 16S rRNA gene reference sequences retrieved from the NCBI and SILVA databases, whereas the remaining ASVs are colored according to their respective class or phylum. Because not all ASVs were resolved to the species or genus level, all ASVs were renamed to a family level for consistency. ASV overlaps between data sets are indicated through the heatmap squares. The maximum-likelihood tree was constructed using a GTR +G +I model with bootstrap support of 1,000.