Oxygen respiration and polysaccharide degradation by a sulfate-reducing acidobacterium

Abstract

Sulfate-reducing microorganisms represent a globally important link between the sulfur and carbon cycles. Recent metagenomic surveys expanded the diversity of microorganisms putatively involved in sulfate reduction underscoring our incomplete understanding of this functional guild. Here, we use genome-centric metatranscriptomics to study the energy metabolism of Acidobacteriota that carry genes for dissimilation of sulfur compounds in a long-term continuous culture running under alternating anoxic and oxic conditions. Differential gene expression analysis reveals the unique metabolic flexibility of a pectin-degrading acidobacterium to switch from sulfate to oxygen reduction when shifting from anoxic to oxic conditions. The combination of facultative anaerobiosis and polysaccharide degradation expands the metabolic versatility among sulfate-reducing microorganisms. Our results highlight that sulfate reduction and aerobic respiration are not mutually exclusive in the same organism, sulfate reducers can mineralize organic polymers, and anaerobic mineralization of complex organic matter is not necessarily a multi-step process involving different microbial guilds but can be bypassed by a single microbial species.

Fig. 1. Bioreactor performance and establishment of a stable enrichment culture over 211 days of operation.

Time-resolved concentration dynamics of sulfate and acetate (A). White and grey shaded areas represent anoxic (0% air-O₂ saturation) and oxic (50% air-O₂ saturation) conditions, respectively. Other organic acids such as malate, fumarate, succinate, citrate, lactate, propionate and butyrate were under the detection limit. Sulfate reduction rates (SRR) were calculated for periods with linear sulfate consumption. SRR were 104 µmol SO₄²⁻ l⁻¹ day⁻¹ (day 113-122), 68 µmol SO₄²⁻ l⁻¹ day⁻¹ (day 144–155) and 65 µmol SO₄²⁻ l⁻¹ day⁻¹ (day 182-191). Alpha diversity (B), beta diversity (C) and microbial community composition at phylum level (D) as determined by 16S rRNA gene amplicon sequencing. Samples for the 16S rRNA gene survey were sequenced in triplicates for most of the time points (filled symbols in B). Numbers within the NMDS plot (C) indicate the day of sampling. Samples for metagenome and -transcriptome sequencing were taken on days 172 (blue arrow) and 185 (red arrow). Source data are provided as Source Data file.

Fig. 2. Phylogenomic reconstruction and metabolic potential of 19 Acidobacteriota metagenome assembled genomes (MAGs) recovered in this study.

MAGs from this study (bold) that contained dsrAB are highlighted in red. The class is indicated at the left of the phylogenetic tree and family affiliation on the right (according to GTDB taxonomy). Abbreviations are as follows: Pol, Polarisedimenticolia; The, Thermoanaerobaculia; HOL, Holophagaceae; FEB10, uncultured family FEB10; UBA1, uncultured family UBA7541; UBA2, uncultured family UBA7540; BRY, Bryobacteraceae; KOR Koribacteraceae; SBA1, uncultured family SBA1; ACI, Acidobacteriaceae. The tree was constructed based on the concatenated alignment of single copy marker genes extracted with GTDB-Tk. Gene presence and absence was determined by an HMM-based annotation using METABOLIC-C. Presence is indicated by filled squares and gene absence by open squares. Grey squares indicate incomplete gene sets. The asterisk indicate gene presence reported in the manually curated annotation, but the complete genes were not detected using the HMM-based annotation. Gene abbreviations and functions are summarized in Supplementary Data 7. Normalized read and transcript abundance is given in reads per kilobase million (RPKM) and indicate the relative abundance in the metagenome and activity in the transcriptome, respectively. Source data are provided as Source Data file.

Fig. 3. Transcriptional activity of Acidobacteriota MAG CO124.

Differentially expressed genes between oxic and anoxic conditions (A). Each dot in the plot indicates a transcriptionally active gene in the genome. Genes that significantly changed in their transcriptional levels according to the Wald test as implemented in DESeq2 (P value adjusted < 0.05) were highlighted in blue. P-values were corrected for multiple testing (Benjamini-Hochberg). Selected hallmark genes of aerobic respiration (ctaDE, petACD) and sulfate reduction (sat, aprAB and dsrAB) are highlighted as triangles. Normalized transcriptional activity (RPKM) of Dsr pathway genes and genes encoding the aerobic respiratory chain (B). Bar charts show the mean ± SD of four technical replicates. Asterisks indicate genes that significantly changed (P value adjusted < 0.05) their transcriptional response between oxic and anoxic conditions. Significance testing was performed using the Wald test as implemented in DESeq2. P-values were corrected for multiple testing (Benjamini-Hochberg). Gene abbreviations and functions are shown in Supplementary Data 2 and 3. Source data are provided as Source Data file.

Fig. 4. Transcriptional pattern of central genes involved in pectin polysaccharide degradation of Acidobacteriota MAG CO124.

Structure of pectin polysaccharides as adapted after Mohnen (2008), arrows indicate the potential target of the carbohydrate active enzymes encoded in MAG CO124 (A). Normalized transcriptional activity (RPKM) of glycoside hydrolases (GH), polysaccharide lyases (PL) and carbohydrate esterases (CE) that were potentially involved in the degradation of pectin polysaccharide is shown in B. Bar charts show the mean ± SD of four technical replicates. Asterisks indicate genes that significantly changed (P value adjusted < 0.05) their transcriptional response between oxic and anoxic conditions. P-values were corrected for multiple testing (Benjamini-Hochberg). Source data are provided as Source Data file.