Metatranscriptomic and metataxonomic insights into the ultra-small microbiome of the Korean fermented vegetable, kimchi

Abstract

Presently, pertinent information on the ultra-small microbiome (USM) in fermented vegetables is still lacking. This study analyzed the metatranscriptome and metataxonome for the USM of kimchi. Tangential flow filtration was used to obtain a USM with a size of 0.2 μm or less from kimchi. The microbial diversity in the USM was compared with that of the normal microbiome (NM). Alpha diversity was higher in the USM than in NM, and the diversity of bacterial members of the NM was higher than that of the USM. At the phylum level, both USM and NM were dominated by Firmicutes. At the genus level, the USM and NM were dominated by Lactobacillus, Leuconostoc, and Weissella, belonging to lactic acid bacteria. However, as alpha diversity is higher in the USM than in the NM, the genus Akkermansia, belonging to the phylum Verrucomicrobia, was detected only in the USM. Compared to the NM, the USM showed a relatively higher ratio of transcripts related to "protein metabolism," and the USM was suspected to be involved with the viable-but-nonculturable (VBNC) state. When comparing the sub-transcripts related to the "cell wall and capsule" of USM and NM, USM showed a proportion of transcripts suspected of being VBNC. In addition, the RNA virome was also identified, and both the USM and NM were confirmed to be dominated by pepper mild mottle virus (PMMoV). Additionally, the correlation between metataxonome and metatranscriptome identified USM and NM was estimated, however, only limited correlations between metataxonome and metatranscriptome were estimated. This study provided insights into the relationship between the potential metabolic activities of the USM of kimchi and the NM.

Keywords: kimchi; metataxonome; metatranscriptome; tangential flow filtration; ultra-small microbiome.

Figure 1

Relative abundance profiling of microbiome analyzed in the USM and NM. Relative abundance profiling of microbial communities was analyzed at the phylum level (A) and genus level (B). A Venn diagram confirmed the relationship between samples on the species level (C).

Figure 2

Relative abundance profiling of metatranscriptome analyzed in the USM and NM. The relative abundance profiling of transcripts was analyzed in level 1 (A), and the relative abundance profiling of sub transcripts related to “cell wall and capsule,” “stress response,” and “virulence, disease, and defense” was analyzed in level 2 (B). A Venn diagram confirmed the relationship between samples on the functional level (C).

Figure 3

Relative abundance profiling of RNA virome analyzed in USM and NM. Relative abundance profiling of RNA virus in the genus level was analyzed using a heat map (A), and the red color intensities of panels are proportional to the operational taxonomic unit (OTU) abundance (max. 82%). Relative abundance profiling of species levels related to Tobamovirus, Carlavirus, and Allexivirus was analyzed using pie charts (B).

Figure 4

Heatmap for correlation between metataxonome and metatranscriptome. The metataxonome was at the genus level, and the metatranscriptome was at level 1. The re-sampling by bootstrapping was performed 20 times. ^*, unclassified bacteria; ^**, unclassified bacteria derived from Betaproteobacteria; ^***, unclassified bacteria derived from Clostridiales; ^****, unclassified bacteria derived from Erysipelotrichaceae; ^*****, unclassified bacteria derived from Gammaproteobacteria; ^******, unclassified bacteria derived from Verrucomicrobia subdivision 3.