Chemosynthetic and photosynthetic bacteria contribute differentially to primary production across a steep desert aridity gradient

Abstract

Desert soils harbour diverse communities of aerobic bacteria despite lacking substantial organic carbon inputs from vegetation. A major question is therefore how these communities maintain their biodiversity and biomass in these resource-limiting ecosystems. Here, we investigated desert topsoils and biological soil crusts collected along an aridity gradient traversing four climatic regions (sub-humid, semi-arid, arid, and hyper-arid). Metagenomic analysis indicated these communities vary in their capacity to use sunlight, organic compounds, and inorganic compounds as energy sources. Thermoleophilia, Actinobacteria, and Acidimicrobiia were the most abundant and prevalent bacterial classes across the aridity gradient in both topsoils and biocrusts. Contrary to the classical view that these taxa are obligate organoheterotrophs, genome-resolved analysis suggested they are metabolically flexible, with the capacity to also use atmospheric H₂ to support aerobic respiration and often carbon fixation. In contrast, Cyanobacteria were patchily distributed and only abundant in certain biocrusts. Activity measurements profiled how aerobic H₂ oxidation, chemosynthetic CO₂ fixation, and photosynthesis varied with aridity. Cell-specific rates of atmospheric H₂ consumption increased 143-fold along the aridity gradient, correlating with increased abundance of high-affinity hydrogenases. Photosynthetic and chemosynthetic primary production co-occurred throughout the gradient, with photosynthesis dominant in biocrusts and chemosynthesis dominant in arid and hyper-arid soils. Altogether, these findings suggest that the major bacterial lineages inhabiting hot deserts use different strategies for energy and carbon acquisition depending on resource availability. Moreover, they highlight the previously overlooked roles of Actinobacteriota as abundant primary producers and trace gases as critical energy sources supporting productivity and resilience of desert ecosystems.

Fig. 1. Microbial community composition, diversity, and abundance of biocrust and topsoil samples collected across the Israel aridity gradient.

Results are shown for biocrust and topsoil samples collected for each climatic zone. a Study site showing satellite imagery basemap with country administrative borders (black dashes) and site geolocations. b Images depict (from left to right) typical site characteristics, biocrust appearance during sample collection, and soil sample appearance following simulated rainfall. c Stacked barchart showing class-level bacterial and archaeal composition based on metagenomic reads of the 16S rRNA gene. Taxonomic classification follows GTDB taxonomy. d Boxplot showing the richness estimate (Chao1) of soil and biocrust communities along the aridity gradient based on 16S rRNA gene amplicon sequencing. e Beta diversity (Bray-Curtis) PCoA ordination visualizing differences in soil and crust community composition between the four climatic zones based on 16S rRNA gene amplicon sequencing. f Boxplot showing 16S rRNA gene copy number for both crust and topsoil along the aridity gradient. Community composition based on 16S rRNA gene amplicon sequencing and richness/beta diversity based on metagenomic sequencing are shown for comparison in Fig. S2.

Fig. 2. Distribution of energy and carbon acquisition genes in biocrust and topsoil microbial communities sampled along the aridity gradient.

a Heatmap showing the abundance of metabolic marker genes in the metagenomic short reads. The percentage of the total community predicted to encode at least one of each gene for a process is shown, based on normalization to single-copy marker genes. b Dot plot showing the metabolic potential of the 68 metagenome-assembled genomes (MAGs). The size of each point represents the number of genomes in each phylum that encode the gene of interest and the shading represents the average genome completeness.

Fig. 3. Maximum likelihood radial phylogenetic trees showing sequence diversity and taxonomic distribution of enzymes responsible for H₂ oxidation and carbon fixation.

a Phylogenetic tree of [NiFe]-hydrogenase large subunit amino acid sequences, with a focus on the group 1h (HhyL) and 1l (HylL) high-affinity uptake hydrogenases to which most binned and unbinned sequences affiliated with. b Phylogenetic tree of RuBisCO large subunit amino acid sequences (RbcL), with a focus on the type IA (Acidimicrobiia-affiliated), type IB (Cyanobacteria-affiliated), and type IE (Actinobacteria- and Thermoleophilia-affiliated) enzymes that most binned and unbinned sequences grouped with. Trees show hits to genome bins (red) and unbinned contigs (blue) relative to reference amino acid sequences (grey) (color figure online).

Fig. 4. Biogeochemical activity measurements of photosynthetic and hydrogenotrophic processes in biocrusts and topsoils collected along the aridity gradient.

a Rates of H₂ oxidation under dry and wet conditions based on gas chromatography. Six biologically independent biocrust and topsoil samples were incubated per climatic zone. b Rates of carbon fixation in wet biocrust and topsoil samples based on tracing ¹⁴C-labelled CO₂ incorporation. Three processes are shown: dark carbon assimilation (i.e., basal rate of CO₂ incorporation under dark ambient conditions due to carbon fixation or anaplerotic processes); photosynthetic carbon fixation (i.e., amount of additional CO₂ fixed under light ambient conditions); and hydrogenotrophic CO₂ fixation (amount of additional CO₂ fixed under dark H₂-enriched conditions). For both biocrusts and topsoils, six biologically independent samples were pooled for each climatic zone and experiments were performed in technical triplicates. For both panels, rates are normalized to 16S rRNA gene copy number as a proxy for biomass (unadjusted rates shown in Fig. S9). Boxplots show median, upper and lower quartile, and minimum and maximum values.

Fig. 5. Conceptual infographic of energy conservation and carbon acquisition detected based on desert metagenome-assembled genomes.

The key enzymes and taxa mediating organic carbon oxidation, light harvesting, trace gas oxidation, and carbon fixation are shown, as are the niches and conditions that these processes are predicted to be most active in.