Actin depolymerizing factor4 is required for actin filament turnover and pattern-triggered immunity in Arabidopsis

Jessica L Henty, Purdue University


The cytoskeleton represents a dynamic platform for detecting and reacting to a diverse collection of biotic and abiotic stimuli. It also comprises a network that is commonly hijacked by intracellular pathogens for their own purposes. As a result, the nature and timing of the changes in actin arrays range from an increase in filament abundance, to extensive bundling, to massive depolymerization, depending on the responding cell type and the nature of signal received. For example, following inoculation with symbiotic bacteria, the host actin cytoskeleton exhibits an increase in filament abundance, whereas infection with pathogenic fungi or oomycetes stimulates an increase in the extent of filament bundling. Many believe that these latter changes to actin filament architecture result from the physical penetration of cells by invading pathogens. Further, studies investigating the actin changes during bacterial pathogenesis, like Pseudomonas syringae infection, which does not penetrate cells, are lacking. Finally, the exact mechanisms used to produce changes in actin filament organization during plant infection are unknown; however, an increase in actin polymerization, a decrease in actin filament turnover, or a combination of the two may be involved. ^ As key signal transducers, numerous actin-binding proteins are known to regulate actin networks, and their response to the common second messengers associated with innate immunity is well characterized in vitro. One actin-binding protein family whose activities are regulated in vitro by pH changes, phospholipids, and protein phosphorylation is the actin depolymerizing factor (ADF)/cofilin family. ADF/cofilin contributes to stochastic filament severing and facilitates actin turnover. However, the actual mechanism of ADF-mediated turnover in cells has not been confirmed. We used powerful reverse-genetics approaches associated with the model organism Arabidopsis thaliana and state-of-the-art microscopy to investigate the loss of an ADF/cofilin isoform on actin filament dynamics in vivo. Epidermal cells from the adf4 homozygous mutant had up to a 3-fold reduction in severing frequency compared to wild-type, as well as increased filament lengths and lifetimes. Collectively, our findings are the first direct evidence for the importance of ADF/cofilin-mediated stochastic severing contributing to the disassembly of actin filament arrays in living cells. ^ We also investigated whether actin rearrangements in the plant host cytoskeleton occur following bacterial treatment using the Arabidopsis-Pseudomonas pathosystem. Using robust actin architecture analysis tools and spinning disk confocal microscopy, we quantified the temporal response of cotyledon epidermal cells to pathogenic and non-pathogenic P. syringae pv. tomato DC3000 strains. We measured an immediate but transient increase in actin filament abundance that was associated with innate immunity. Additionally, we used the genetics of both bacterium and plant to demonstrate that the increase in actin filament abundance still occurred in epidermal cells inoculated with non-pathogenic mutants deficient for type-III secretion or for all 28 putative effector proteins. Significantly, a similar actin response was elicited with treatments of bacterial and fungal microbe-associated molecular patterns (MAMPs). We also observed an increase in the extent of actin filament bundling at late timepoints, which appeared to be part of host-plant susceptibility. This is the first evidence that the host-cell cytoskeleton responds to multiple signals at different times following treatment with pathogenic bacteria. Finally, several Arabidopsis knockout mutants for a well-characterized MAMP receptor complex failed to elicit a change to actin arrays following DC3000 inoculation. Collectively, these results demonstrate unambiguously that the perception of MAMP ligands and receptor kinase signaling are necessary to trigger actin reorganization in host epidermal cells. ^ To further dissect the cellular mechanisms that underpin the increase in actin filament abundance during innate immunity, we utilized the Arabidopsis adf4 mutant and the high spatiotemporal resolution imaging of hypocotyl epidermal cells. If ADF4 activity is negatively regulated during innate immunity, we predict that the cytoskeleton in the adf4 knockout mutant will resemble the actin array changes observed during PTI. Furthermore, if ADF4 is a key player that contributes to the increase in actin filament abundance, the knockout mutant should be unresponsive to MAMP signals. The changes we observed following MAMP treatment were consistent with alterations to actin filament turnover and actin filament end dynamics. Indeed, several changes in stochastic dynamics properties phenocopied the features of the actin cytoskeleton in the adf4 mutant measured previously. Moreover, the adf4 mutant does not display changes in actin filament abundance following MAMP treatment. This provides compelling evidence that ADF4 activity is down regulated during PTI, and is the first data implicating a key actin-binding protein in the rapid actin turnover that occurs during plant innate immunity.^




Christopher J. Staiger, Purdue University.

Subject Area

Biology, Botany|Biology, Cell|Agriculture, Plant Pathology

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server