97th ESA Annual Meeting (August 5 -- 10, 2012)

PS 59-178 - Microbial primary successional dynamics: An assessment of community convergence/divergence patterns

Wednesday, August 8, 2012
Exhibit Hall, Oregon Convention Center
Shawn P. Brown, Division of Biology, Kansas State University, Manhattan, KS and Ari Jumpponen, Department of Biology, Kansas State University, Manhattan, KS
Background/Question/Methods

The study of microbial community dynamics in primary successional ecosystems offers unique insight into the processes of community assembly.  Previous studies of soil microbial communities along a successional chronosequence at Lyman Glacier (Glacier Peak Wilderness Area, WA, USA) indicate clear community turnover and dramatic shifts in community composition.  In this study, we aimed to determine if these shifts in bacterial and fungal communities exhibit convergence/divergence when measured by genetic distances based on 454-pyrosequenced PCR amplicons (28S for Fungi, 16S for Bacteria). We targeted environmental DNA from rhizosphere soils collected at 0-750 m distances from the glacier terminus. The samples represent four common plant species with differing mycorrhizal habits: Abies lasiocarpa (Ectomycorrhizal), Luetkia pectinata (Arbuscular Mycorrhizal), Phyllodoce empetriformis (Ericoid Mycorrhizal), Saxifraga ferruginea (Non-mycorrhizal) as well as non-vegetated soil.  Using mothur (v.1.23.0), we clustered sequences (97% similarity), aligned resultant clusters with well-curated references, and generated neighbor-joining (NJ) trees. Weighted UniFrac analyses were used to determine community similarity.  Additionally, nonmetric multidimensional scaling (NMS) was used to visualize community structure and to determine patterns of community composition.

Results/Conclusions

Sequence analysis indicates diverse microbial communities even in a relatively simple primary successional system.  Based on our NMS analysis, bacterial communities distinctly shift over the chronosequence along Axis 1 (F1,55=5.615, P=0.213) and differ among plant rhizospheres along Axis 2 (F4,55=2.731, P=0.0365).  Analysis of molecular variance (AMOVA) tests show that the centroids across all 5 axes of the NMS data cloud differ among many of the vegetation types and distance categories. Across vegetation types (F4,60=2.24, P<0.001), bare soil is distinct from all vegetation rhizosphere samples (P<0.001),  Phyllodoce is distinct from Saxifraga (F1,24=1.97, P=0.003).  Across all distances (F5,59=2.60, P<0.001), recently exposed soils (0 and 150 m) generally differ in community structure from older soils (300, 450, 600, 750 m).  UniFrac analysis suggests a trend of increasing NJ branch length with distance from glacier terminus (t=1.72, P=0.0908) for bacteria. These results suggest community divergence with increasing successional age.  This overall pattern of increased NJ branch length is driven by the distinct increase in Luetkia rhizosphere and bare soil (P=0.0178, P=0.0007 respectively) environments whereas other species appear to possess static communities over the chronosequence. Fungal communities also show distinct community dynamics and differ in community composition across soils varying in distance from the glacier and in vegetation.