Functional redundancy buffers the effect of poly-extreme environmental conditions on southern African dryland soil microbial communities

Abstract

Drylands' poly-extreme conditions limit edaphic microbial diversity and functionality. Furthermore, climate change exacerbates soil desiccation and salinity in most drylands. To better understand the potential effects of these changes on dryland microbial communities, we evaluated their taxonomic and functional diversities in two Southern African dryland soils with contrasting aridity and salinity. Fungal community structure was significantly influenced by aridity and salinity, while Bacteria and Archaea only by salinity. Deterministic homogeneous selection was significantly more important for bacterial and archaeal communities' assembly in hyperarid and saline soils when compared to those from arid soils. This suggests that niche partitioning drives bacterial and archaeal communities' assembly under the most extreme conditions. Conversely, stochastic dispersal limitations drove the assembly of fungal communities. Hyperarid and saline soil communities exhibited similar potential functional capacities, demonstrating a disconnect between microbial structure and function. Structure variations could be functionally compensated by different taxa with similar functions, as implied by the high levels of functional redundancy. Consequently, while environmental selective pressures shape the dryland microbial community assembly and structures, they do not influence their potential functionality. This suggests that they are functionally stable and that they could be functional even under harsher conditions, such as those expected with climate change.

Keywords: dryland soils; edaphic microbial communities; functional redundancy; metabarcoding; niche partitioning; shotgun metagenomics.

Figure 1.

Map of South Africa with the sampling sites. The map shows the distribution of the sampling sites colored according to their dryland of origin (Namaqua and Richtersveld). Representative photographs of each sample site are included (photos courtesy of J.-B. Ramond). Satellite images were produced using © Mapbox.

Figure 2.

Southern African dryland edaphic microbial community diversity. MDS plots of bacterial and archaeal (A), and fungal (B) communities with weighted Unifrac distances calculated over log₁₀-transformed data. The points are colored according to the dryland (NNP, Namaqua; RNP, Richtersveld), and the labels indicate the corresponding sample. Stacked bar charts showing class-level bacterial and archaeal (C), and fungal (D) composition with the respective domain in parentheses. A, Archaea; B, Bacteria.

Figure 3.

Ecological processes driving Southern Africa dryland soil microbial communities assembly. Stacked bar chart showing the contribution of each process to the assembly of the hyperarid, arid, and arid saline soil bacterial and archaeal (A), and fungal (B) communities.

Figure 4.

Abundance and distribution of key functional genes related to microbial stress response in Southern African dryland soils. The seven phyla with the highest gene abundances (>35.2 TPM) in both drylands are shown in the x-axis, whilst less represented phyla are grouped in the “Others” category. Colored asterisks indicate the gene families that were significantly enriched in one of the drylands (two-sided Fischer's exact test; q-value < 0.05). The TPM values are squared-root transformed.

Figure 5.

Abundance and distribution of key functional genes related to microbial metabolism and nutrient cycling in Southern African dryland soils. Genes related to carbon, nitrogen, and sulfur cycling pathways are displayed. The seven phyla with the highest gene abundances (>13.5 TPM) in both drylands are shown in the x-axis. The less represented phyla are grouped in the ‘Others’ category. Colored asterisks indicate the gene groups that were significantly enriched in one of the drylands (two-sided Fischer's exact test; q-value < 0.05). The TPM values are squared-root transformed. mycoS_dep_FDH: mycothiol-dependent formaldehyde dehydrogenase.