Clostridiaceae and Enterobacteriaceae as active fermenters in earthworm gut content

Abstract

The earthworm gut provides ideal in situ conditions for ingested heterotrophic soil bacteria capable of anaerobiosis. High amounts of mucus- and plant-derived saccharides such as glucose are abundant in the earthworm alimentary canal, and high concentrations of molecular hydrogen (H(2)) and organic acids in the alimentary canal are indicative of ongoing fermentations. Thus, the central objective of this study was to resolve potential links between fermentations and active fermenters in gut content of the anecic earthworm Lumbricus terrestris by 16S ribosomal RNA (rRNA)-based stable isotope probing, with [(13)C]glucose as a model substrate. Glucose consumption in anoxic gut content microcosms was rapid and yielded soluble organic compounds (acetate, butyrate, formate, lactate, propionate, succinate and ethanol) and gases (carbon dioxide and H(2)), products indicative of diverse fermentations in the alimentary canal. Clostridiaceae and Enterobacteriaceae were users of glucose-derived carbon. On the basis of the detection of 16S rRNA, active phyla in gut contents included Acidobacteria, Actinobacteria, Bacteroidetes, Chloroflexi, Cyanobacteria, Firmicutes, Gemmatimonadetes, Nitrospirae, Planctomycetes, Proteobacteria, Tenericutes and Verrucomicrobia, taxa common to soils. On the basis of a 16S rRNA gene similarity cutoff of 87.5%, 82 families were detected, 17 of which were novel family-level groups. These findings (a) show the large diversity of soil taxa that might be active during gut passage, (b) show that Clostridiaceae and Enterobacteriaceae (fermentative subsets of these taxa) are selectively stimulated by glucose and might therefore be capable of consuming mucus- and plant-derived saccharides during gut passage and (c) indicate that ingested obligate anaerobes and facultative aerobes from soil can concomitantly metabolize the same source of carbon.

Figure 1

Distribution of RNA in gradient fractions of glucose treatments. Arrows with fraction numbers indicate ‘heavy' (3 and 4) and ‘light' (8 and 9) fractions, which were chosen for further analyses. RNA was measured in all gradient fractions of one replicate. Symbols: empty symbols, RNA extracted at the start of incubation; filled symbols, RNA extracted at 51 h of incubation; circles, [¹²C]glucose treatment; squares, [¹³C]glucose treatment.

Figure 2

Effect of supplemental glucose (a) on the production of CO₂ (b) and H₂ (c). The symbols indicate the means of triplicates; error bars indicate s.d.. Arrows indicate the time when samples were taken for RNA stable isotope probing analyses. Insets in (b) and (c) are enlargements of the initial incubation period. Symbols: ▪, results from [¹³C]glucose treatment; □, results from [¹²C]glucose treatment; ○, results from unsupplemented control.

Figure 3

Rarefaction curves of 16S rRNA complementaryDNA sequences were obtained at the start of incubation (empty circles) and after 51 h of incubation (filled circles). Each curve is based on the total sequences obtained from [¹²C]glucose- and [¹³C]glucose-supplemented incubations. Solid lines enclosing the symbols indicate the 95% confidence intervals.

Figure 4

Phylogenetic tree of 16S rRNA complementaryDNA sequences (bold) retrieved from the earthworm gut and reference sequences of the Firmicutes (a) and Gammaproteobacteria (b). Only representative sequences are shown. The values next to the branches represent the percentages of replicate trees (>50%) in which the associated taxa clustered together in the bootstrap test (10 000 replicates). Dots at nodes indicate confirmation of topology by AxML using the same data set. Labeled taxa are marked with stars. Methanosarcina barkeri (AF028692) was used as outgroup. Accession numbers are in parentheses. Quotation marks indicate nonvalidated taxa (Euzéby, 2010). Abbreviations: t₀, at the start of incubation; t₅₁, at 51 h of incubation; L, ‘light' fractions; H, ‘heavy' fractions. (a): The family Veillonellaceae was excluded because the sequence that affiliated with this family (t51_12H90) was shorter than 700 bp. The phylogenetic tree was calculated using the neighbor-joining method (50% minimum similarity filter; 651 valid positions between 104 and 815 of the 16S rRNA gene of E. coli). Bar indicates 0.01 estimated change per nucleotide. (b): The labeled sequence t51_13H112 (next cultivated species Serratia fonticola, AY236502, 99% 16S rRNA gene similarity) was excluded from tree calculation because it was shorter than 700 bp. The phylogenetic tree was calculated using the neighbor-joining method (50% minimum similarity filter; 698 valid positions between 101 and 816 of the 16S rRNA gene of E. coli). Bar indicates 0.05 estimated change per nucleotide.

Figure 4

Phylogenetic tree of 16S rRNA complementaryDNA sequences (bold) retrieved from the earthworm gut and reference sequences of the Firmicutes (a) and Gammaproteobacteria (b). Only representative sequences are shown. The values next to the branches represent the percentages of replicate trees (>50%) in which the associated taxa clustered together in the bootstrap test (10 000 replicates). Dots at nodes indicate confirmation of topology by AxML using the same data set. Labeled taxa are marked with stars. Methanosarcina barkeri (AF028692) was used as outgroup. Accession numbers are in parentheses. Quotation marks indicate nonvalidated taxa (Euzéby, 2010). Abbreviations: t₀, at the start of incubation; t₅₁, at 51 h of incubation; L, ‘light' fractions; H, ‘heavy' fractions. (a): The family Veillonellaceae was excluded because the sequence that affiliated with this family (t51_12H90) was shorter than 700 bp. The phylogenetic tree was calculated using the neighbor-joining method (50% minimum similarity filter; 651 valid positions between 104 and 815 of the 16S rRNA gene of E. coli). Bar indicates 0.01 estimated change per nucleotide. (b): The labeled sequence t51_13H112 (next cultivated species Serratia fonticola, AY236502, 99% 16S rRNA gene similarity) was excluded from tree calculation because it was shorter than 700 bp. The phylogenetic tree was calculated using the neighbor-joining method (50% minimum similarity filter; 698 valid positions between 101 and 816 of the 16S rRNA gene of E. coli). Bar indicates 0.05 estimated change per nucleotide.