Original paper

Biodiversity and ecology of soil fungi in a primary succession of a temperate coastal dune system

Prenafeta Boldú, Francesc X.; Summerbell, Richard C.; de Boer, Wietse; Boschker, Henricus T.S.; Gamsformerly, Walter

Nova Hedwigia Band 99 Heft 3-4 (2014), p. 347 - 372

published: Nov 1, 2014

DOI: 10.1127/0029-5035/2014/0203

BibTeX file

ArtNo. ESP050009903002, Price: 29.00 €

Download preview PDF Buy as PDF


Soil fungal communities were studied in an actively developing coastal dune system at Goeree Island, the Netherlands. A shore to inland sampling transect was laid out, extending from coastal brackish marshes to recently formed foredunes to older dune pastures to adjacent woodlands. Soil samples from these biotopes were thoroughly characterized by analyzing physicochemical and microbial characteristics. Soil fungal community structure and composition were analysed by a combination of different phenotypic and genotypic methodologies (isolation of microfungi via a specialized soil washing technique and in situ observation of macrofungi, versus DGGE profiling and sequencing of multi-locus rDNA clone libraries). The results showed that fungal biomass tended to increase land-inwards along the gradient of maturity. The community structure was significantly correlated with progressive soil acidification land-inwards and with the exposure to brackish water in the coastal sites. Comparison between isolation and molecular datasets revealed that both methods were biased towards specific functional or phylogenetic groups. Most of the isolated fungi were common soil saprotrophic ascomycetes, while specialized fungi (biotrophic plant symbionts and pathogens, primary decomposers of recalcitrant organic matter, etc.) were only detected by molecular means. Phylogenetic specificity of PCR-based DNA profiling, on the other hand, strongly depended on primer selection. In spite of the relatively low number of common species that were identified among the isolated cultures and by clone library sequencing, as well as the potential biases of each characterization method, multivariate analysis on both isolation and molecular datasets yielded similar correlation patterns with the environment.