Figure 1.
Proportion of overall phyla and disectioning of the fungal and Glomeromycota phyla detected by pyrosequencing analysis with primers NS31/AML2 from rhizosphere samples obtained from 96 olive orchards in Andalusia, southern Spain.
Table 1.
Glomeromycota taxa detected in olive rhizosphere samples, taxonomic affiliation, number of sequences from each taxa, frequency of occurrence in the sampled olive orchards, and the closest related AMF sequence.
Figure 2.
Phylogenetic relationships of nuclear small subunit ribosomal RNA (SSU rRNA) gene sequences of Glomeromycota reference sequences derived from uclust_ref search with those that matched the silva_108 and MaarjAM databases, the reference AMF database from Redecker and Raab [28] and those reported in olive from Calvente et al [7] and present in the GenBank.
Bayesian 50% majority rule consensus tree as inferred from nSSU rRNA sequences alignments under the general time reversible + G + I model. Numbers on the nodes indicate Bayesian posterior probabilities (>50%). The phylogram was rooted with Paraglomeraceae sequences. Numerical codes in bold name each representative AMF OTUs from olive rhizosphere derived from uclust_ref analysis with the QIIME software and are labelled (OAMF# S#) where OAMF refers to ‘olive arbuscular mycorrhizal fungi’ and # to the number assigned and groups identified and the remaining code refers to the soil sample. Phylogenetic groups (I to Xd) were arbitrarily described and are shown in Table 1. (*) Although this sequence was originally identified as belonging to R. clarus by Calvente et al. [7] its closest taxonomic affiliation is to Funneliformis sp. and clearly differs from sequences AJ276084 and AJ852597 of R. clarus.
Figure 3.
NMDS biplot of a Bray-Curtis dissimilarity matrix of T-RFLP analysis.
The fitted vectors of environmental and physicochemical soil variables and the agronomic variable age of plantation (indicated with different symbols) most significantly and strongly associated (P<0.05) with the ordination and shown in Table 2 are also represented. Size of symbols is proportional to AMF richness in those olive orchards.
Table 2.
Summary of relationshipsa between agronomic, soil and environmental factors and AMF communities assessed by T-RFLP analysis.
Figure 4.
Sums of squares multivariate regression tree summarizing olive AMF community–agronomic, environmental and soil factors relationships.
The tree was calculated using frequency of AMF OTUs derived from pyrosequencing analysis (). For each split a rule is selected based on the predictors to minimize the dissimilarity within the AMF OTUs profiles in the resulting two nodes (main rule is shown above the node, and second rules are shown below the node). At each terminal node, the mean relative abundances of each AMF OTU are shown, together with the number of olive orchards for each group.