Untangling the Food Web Inside Broom Galls
Broom inoculation experiment.
Gall mites (Aceria genistae) were introduced to New Zealand from France and released in 2008 to damage broom (Cytisus scoparius) stems and reduce their vigour. When biocontrol agents are established it is inevitable they will interact with many more organisms than just their target weeds as they become part of new food webs. In Issue 57 we reported finding that the gall mites were carrying microbial spores. At the time, it was not known whether these microbes were fungi or bacteria and whether they were helping or hindering the activity of the gall mite. Researchers wanted to explore whether there is a synergistic relationship between the microbes and the gall mites that is important in gall formation, and therefore whether they play a role in regulating the growth of the broom. “We are always interested in learning what makes some agents more successful than others so we can make better agent choices in the future,” said Zhi- Qiang Zhang, the acarologist leading this project.
Several years down the track, and after input from a range of scientific disciplines, we have learnt that there are a diverse range of microbes associated with the broom galls and gall mites in New Zealand and similar genera in France. Chantal Probst collected galled stems from broom plants in both countries. Using DNA analysis she isolated more than 80 genera of fungi, yeast and bacteria from the New Zealand samples and almost 50 genera from the French samples (which were also fewer in number), a much higher diversity than expected. Pinpointing the origin of the microbes found on the New Zealand plants, although desirable from a biosecurity perspective, was not possible since there is no historical record of which microbes occur naturally in New Zealand to use as a reference.
The next question was whether the microbes were pathogenic to the broom plants. Experiments were designed to determine if the common microbes could form galls themselves, in the absence or presence of the gall mites. “Secondary to this, we needed to determine whether the microbes affected the growth and survival of the plant in any way,” explained Zhi-Qiang. This work was done in two phases: firstly in glasshouses and then using field studies based in North Canterbury. In phase one young broom plants were inoculated with fungi (two species of Fusarium and one species of Phoma) and one species of bacterium (Pantoea) extracted from broom galls. These were compared with three different controls, which included a positive inoculation with a known pathogen (Fusarium tumidum). There were five replicates of each treatment and the experiment was repeated three times. The results indicated that, after 6 weeks, there was no difference in the growth rate of the inoculated plants and the control plants that were untreated. There were no galls visible either, suggesting that the isolates were not able to induce gall formation in the absence of the mite within this time frame.
Phase two is currently underway at a trial site on Leslie Hill Station near Hanmer Springs. At this site, the number of fungal isolates, the abundance of gall mites and the number of predatory mites (that reduce the number of gall mites) are being manipulated. Early results indicate that the gall mites are doing a good job of inducing galls on the broom, which is leading to lower plant survival. The broom gall mite is under attack from predatory mites, but broom plants treated with a miticide that specifi cally targets predatory mites (therefore allowing the gall mite populations to increase in abundance) did not appear to be adversely affected in terms of survival. Despite the predatory mites, the broom gall mites are still able to perform well. The galls offer some protection to the broom gall mites, and it appears that predatory mite numbers build up too late in the season to have a major impact. However, if mites are to be shifted to new sites, establishment success is likely to be greater if done early in the season (October–December) before predator numbers build up. Again no galls have been formed during the 2 years that the field experiment has been running, except by the gall mites, backing up the lab results suggesting that microbes are not involved in gall formation.
“The overall food web associated with broom galls has proved quite diffi cult to untangle,” said Zhi-Qiang. We have confirmed that there are fungivore–fungi–plant interactions and predatory–prey–plant interactions going on, in what is known as a reticulated trophic web, linked by polyphagous mites (see graphic). “There are still many complexities to unravel before we are able to determine the true relationship between the microbes and the gall mites,” remarked Quentin Paynter. It is possible that the microbes only attach themselves to the mites once they have been blown onto the plant and they are literally hitching a lift on the body of the mite to the part of the plant that they can infect. This was supported by the results shown from the glasshouse experiments and would defi nitely mean that the relationship between the microbes and the mites has an important benefi cial effect on broom control. A logical extension to this research would be to confirm the role of the microbes found, but given their diversity, this could take a considerable amount of time and funding, so opportunities for students to undertake this research are being explored.
This research was funded by the Ministry of Business, Innovation and Employment as part of Landcare Research’s Beating Weeds Programme and Capability Funding.