|Stein, Ilana -|
Submitted to: Biological Invasions
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: January 29, 2012
Publication Date: February 1, 2012
Citation: Swope, S.M., Stein, I.R. 2012. Soil type mediates interactions between a plant, its pathogen and seed predators. Biological Invasions. 14:1697-1710. Interpretive Summary: Invasive plants pose one of the greatest threats to biodiversity. For many of the most widespread and well-established invaders, biological control offers us our best chance to reduce their abundance. Biological control (aka biocontrol) involves the importation and release of the invader’s enemies from its native range with the idea that their attack will reduce the abundance of the invader or the rate at which it is spreading into uninvaded areas. Here we focused on the impact of a suite of biocontrol agents on one of California’s most problematic and widespread invasive plants, yellow starthistle, in one of California’s most important conservation areas, serpentine soils. Serpentine soils occur in areas of tectonic activity and provide abiotic conditions that most plants find very stressful. As such, few plants can survive on these soils, but those that can have evolved into distinct (endemic) species. Serpentine soils tend to occur in discrete patches within a matrix of otherwise benign soil types which are often heavily invaded by exotic plants, including yellow starthistle. So, not only do serpentine soils support a diverse community of endemic plants (and other taxa), which contribute to biodiversity in a unique way, but they also provide a refuge from invasion on the surrounding soils. One of the characteristics that make serpentine soils so stressful to plants is their low calcium content. Calcium plays a critical role in allowing the plant to both detect and defend itself against pathogen attack so we hypothesized a newly released biocontrol pathogen (yellow starthistle rust) might have a larger effect on yellow starthistle plants growing on serpentine soil. Both the stressful serpentine soils and the pathogen reduced starthistle performance (although the impact of the pathogen was not larger on the serpentine soils than it was on adjacent non-serpentine soil). Additionally, the pathogen made the plant more vulnerable to seed predation by established seed-feeding insect biocontrol agents on the serpentine soil (but not on the non-serpentine soil) and the net effect of these many interactions was that the pathogen reduced the number seeds plants produced by half on the serpentine soils. Although this pathogen appears only modestly impact plants elsewhere, our data show that it has the potential to significantly reduced starthistle on serpentine soils that are of high conservation concern in California.
Technical Abstract: Plants interact with numerous enemies and mutualists over the course of their lifetimes, potentially giving rise to indirect interactions and non-additive outcomes. This topic is central to the use of biocontrol agents to manage invasive plants because the current approach is to release multiple agents for each weed species. Previously, we demonstrated that a biocontrol pathogen had a direct, negative effect on its host, the invasive plant Centaurea solstitialis, but also had a positive impact by indirectly reducing seed predation. Here we extend our previous findings by identifying the mechanism and exploring how these interactions change in response to abiotic conditions. We hypothesized that pathogen-induced resistance to seed predation may be due to systemic acquired resistance (SAR) or a change in the plant’s nutrient status. Serpentine soils are of particular conservation concern in California because they support numerous endemic species. They also have very low calcium concentrations and Ca is essential to a plant’s ability to respond to pathogen infection. If SAR is important, serpentine plants may be more susceptible to the pathogen’s direct effects and less defended against seed predators. We experimentally infected C. solstitialis plants on serpentine and non-serpentine soil and found that both serpentine soil and pathogen infection reduced plant performance although the pathogen’s direct impact was not greater on serpentine plants. We also found that even when infected plants produced seeds with higher nitrogen content, all four species of seed predators consistently ate fewer seeds when the plant was infected than when it was healthy. In at least one case, infected serpentine plants were less defended against seed predators than infected non-serpentine plants. This supports SAR as the mechanism. On the non-serpentine soil, the pathogen’s negative impact was entirely canceled out by its indirect, positive effect. On the serpentine soil, pathogen infection reduced lifetime seed output by half. This combination of biocontrol agents had a large impact on lifetime fitness of plants on serpentine soils with high conservation value. More broadly, our results demonstrate that abiotic conditions may strongly influence the relative strength of indirect interactions.