96th ESA Annual Meeting (August 7 -- 12, 2011)

COS 11-5 - How the seed bank pathogen Pyrenophora semeniperda kills non-dormant cheatgrass seeds

Monday, August 8, 2011: 2:50 PM
12B, Austin Convention Center
Heather Finch, Plant and Wildlife Sciences, Brigham Young University, Provo, UT, Susan E. Meyer, Shrub Sciences Laboratory, USDA Forest Service, Rocky Mountain Research Station, Provo, UT and Phil S. Allen, Department of Plant and Wildlife Science, Brigham Young University, Provo, UT
Background/Question/Methods

Cheatgrass (Bromus tectorum) is an invasive annual grass that occurs as a monoculture over millions of hectares in the semiarid western United States.  The seed bank pathogen black fingers of death (Pyrenophora semeniperda) can kill thousands of cheatgrass seeds per square meter each year in these monoculture stands. Laboratory inoculation experiments with fully-hydrated cheatgrass seeds have shown that dormant seeds suffer high mortality, whereas non-dormant seeds generally escape through rapid germination. This led us to believe that the primary prey of this pathogen in the field is the secondarily dormant spring seed bank. However, field inoculation trials have indicated that a sizeable fraction of the non-dormant seed bank can also be killed.  We hypothesized that the pathogen could kill non-dormant seeds at suboptimal water potentials restrictive to seed germination.  To test this hypothesis, non-dormant and dormant cheatgrass seeds were inoculated with pathogen conidia and pre-incubated at 20⁰ C for 28 days over a range of water potentials from 0 to -2.0 MPa.  At days 2, 4, 7, 11, 14, 21 and 28, seeds were scored as ungerminated, germinated or killed (pathogen stromata present).  After 28 days, seeds were switched to free water (0 MPa) and incubated and scored for an additional 28 days. 

Results/Conclusions

Experiments with dormant seeds showed that >90% of seeds were killed regardless of treatment.  The fraction of seeds that produced stromata at reduced water potentials was lower due to the delayed time course of disease development.  Germination rates for non-dormant seeds were also significantly slowed as water potential was reduced.  By day 28, seeds pre-incubated at -2.0 MPa showed only 3% germination compared to seeds at -0.5 MPa which showed 99% germination.  Experiments with non-dormant seeds demonstrated that, at less negative water potentials, few or no seeds were killed due to fast germination. At lower water potentials, the fraction of ungerminated seeds showing stromata increased dramatically during the 28-day incubation period in free water.  After pre-incubation at -1.5 MPa, 67% of non-dormant seeds showed stromata by day 7.  Similarly, after pre-incubation at -2.0 MPa, 74% of non-dormant seeds showed stromata by day 14.  We conclude that this pathogen can infect and kill cheatgrass seeds at water potentials below the threshold for germination of non-dormant seeds.  This is a plausible explanation as to how non-dormant seeds can be killed in the field.