The core bacteriobiome of Côte d'Ivoire soils across three vegetation zones

Abstract

The growing understanding that soil bacteria play a critical role in ecosystem servicing has led to a number of large-scale biogeographical surveys of soil microbial diversity. However, most of such studies have focused on northern hemisphere regions and little is known of either the detailed structure or function of soil microbiomes of sub-Saharan African countries. In this paper, we report the use of high-throughput amplicon sequencing analyses to investigate the biogeography of soil bacteria in soils of Côte d'Ivoire. 45 surface soil samples were collected from Côte d'Ivoire, representing all major biomes, and bacterial community composition was assessed by targeting the V4-V5 hypervariable region of the 16S ribosomal RNA gene. Causative relationships of both soil physicochemical properties and climatic data on bacterial community structure were infered. 48 phyla, 92 classes, 152 orders, 356 families, and 1,234 genera of bacteria were identified. The core bacteriobiome consisted of 10 genera ranked in the following order of total abundance: Gp6, Gaiella, Spartobacteria_genera_incertae_sedis, WPS-1_genera_incertae_sedis, Gp4, Rhodoplanes, Pseudorhodoplanes, Bradyrhizobium, Subdivision3_genera_incertae_sedis, and Gp3. Some of these genera, including Gp4 and WPS-1_genera_incertae_sedis, were unequally distributed between forest and savannah areas while other taxa (Bradyrhizobium and Rhodoplanes) were consistently found in all biomes. The distribution of the core genera, together with the 10 major phyla, was influenced by several environmental factors, including latitude, pH, Al and K. The main pattern of distribution that was observed for the core bacteriobiome was the vegetation-independent distribution scheme. In terms of predicted functions, all core bacterial taxa were involved in assimilatory sulfate reduction, while atmospheric dinitrogen (N₂) reduction was only associated with the genus Bradyrhizobium. This work, which is one of the first such study to be undertaken at this scale in Côte d'Ivoire, provides insights into the distribution of bacterial taxa in Côte d'Ivoire soils, and the findings may serve as biological indicator for land management in Côte d'Ivoire.

Keywords: 16S rDNA; Côte d’Ivoire; bacteria; common core bacteriobiome; microbiome.

Figure 1

Map of Côte d’Ivoire showing the 45 sampled sites. Geographic coordinates of the sampled sites are given in Supplementary Table S1. The map is colored according to the main vegetation types in Côte d’Ivoire (forest, savannah and forest-savannah transition zones).

Figure 2

Rarefaction curve of the forty-five samples showing the number of ASVs recorded as a function of sequences count.

Figure 3

Classification of bacterial phyla from Côte d’Ivoire soils according to their abundance and relative occurrence. The occurrence varies between 100% for a total occupation of the soils and about 2% for a presence in a single soil.

Figure 4

Contribution and effects of each variable (geographical location, climate and physico-chemical properties) in the distribution of bacterial taxa. The proportions of variance explained were the significant contributions of these variables (p < 0.05). The size of the circles as well as the values written inside provide information on the proportion of variance explained by each variable. The colors blue and red indicate the effect of the regression coefficients based on Spearman’s correlation (blue, positive effect, red, negative effect).

Figure 5

The core bacteriobiome of Côte d’Ivoire soils: taxonomic distribution per sample for the most abundant genera with 100% occurrence. Hierarchical clustering of samples based on the relative abundances of the core genera is shown by the dendrogram on the bottom of the plot which revealed three distinct groups: Gr1, Gr2, and Gr3.

Figure 6

Redundancy analysis (RDA) ordination diagram (triplot) of core bacteria genera and physicochemical factors across Côte d’Ivoire soils. The plot shows samples in dots, explanatory physicochemical variables (red arrows), and the response variables (orange square with genera names). Axis 1 and axis 2 explained 37.9 and 9.6% of the total variation, respectively. Positively correlated variables are shown as arrows going in the same direction.

Figure 7

PICRUSt2 predicted metabolic capacities of the core genera based on 16S rRNA genes and the KEGG database. Abundance values are ranked from the lowest (light blue) to the highest (red). The grey boxes means no value recorded.

Figure 8

Relative abundance of genes associated with agriculturally important energy metabolism functions in the core bacteriobiome of Côte d’Ivoire soils.

Figure 9

Distribution of module abundance within each energy metabolism function for each core genus. The plot was split regarding the main Energy metabolism functions. No bar means that there was no data for the concerned genus.