Using qPCR to correlate the abundance of diazotrophic soil bacteria with the abundance of understory plants in a Douglas-fir (Pseudotsuga menziesii) forest on Vancouver Island

Background/Question/Methods Nitrogen-cycling bacteria are of particular interest in forestry, because the soils of many temperate forests are nitrogen-limited; it is one factor that affects sustainability.  At the LOGS (Levels of Growing Stock) experiment in British Columbia, Douglas-fir forest plots were used to study the impact of understory plants on populations of diazatrophic (nitrogen-fixing) soil bacteria.  In previous work at the LOGS site, a quantitative polymerase chain reaction (qPCR) method was developed to quantify soil bacteria possessing the nifH (nitrogen reductase) gene, using universal primers with nested PCR, as well as primers specific to free-living diazotrophic bacteria (Azotobacter spp.).  In the current study, the abundance of these bacteria was compared with the abundance of understory plant species with potential impacts on the rhizosphere.  In 2007, the percent cover of understory plants was visually estimated in nine 0.08 ha LOGS plots near Shawnigan Lake, on Vancouver Island.  Linear regression analysis was used to compare the cover data versus the log of nifH gene copies (universal and specific primers) measured in the previous work, averaging the numbers obtained for organic and mineral soil horizons.

Results/Conclusions Understory plants in the plots were categorized as salal (Gaultheria shallon), Oregon grape (Mahonia aquifolium), bracken fern (Pteridium aquilinum), sword fern (Polystichum munitum), and moss species (as a group).  The population of bacteria detected by the universal nifH method did not significantly correlate with changes in the percent cover of salal (P=0.21), with measurements remaining at ca. 10⁶-10⁷ gene copies per gram of soil.  However, abundance of Azotobacter spp. correlated significantly (P=0.016, R²=0.40) with salal, maintaining ca. 10³-10⁴ gene copies per gram of soil in sparse salal, but almost becoming absent  in high densities of the plant.  It also correlated significantly (P=0.0005, R²=0.69) with moss, becoming more abundant as the percentage of moss cover increased.  The abundance of moss was inversely correlated with the abundance of salal (P=0.000007, R² = 0.87).  There were no significant correlations vs. the other understory plants.