W.R. Bushnell

Hypersensitive cell death, autofluorescence, and insoluble silicon accumulation in barley leaf epidermal cells under attack by Erysiphe graminis f. sp. hordei***

Hypersensitive cell death (HR) of adaxial leaf epidermal cells of barley containing the M1a gene for resistance to powdery mildew germlings, Erysiphe graminis f. sp. hordei , was investigated for temporal relations between cell death, onset of autofluorescence, and accumulation of insoluble silicon (Si). Physiological cell viability or death of barley epidermal cells was determined by concomitant uptake of the vital dye neutral red, and by cell plasmolysis using alpha methyl- d -glucose. Cell death, as judged by lack of neutral red uptake and lack of plasmolysis, first occurred in a low percentage of attacked epidermal cells at 15 h after inoculation and increased to a maximum of 72% at 21 and 24 h. Cells judged dead by physiological criteria always exhibited whole-cell autofluoresence by blue light excitation. Autofluorescence was weak in dying and recently dead cells, but it increased in intensity with time following cell death. It was concluded that autofluorescence accompanies cell death and is perhaps due to release of phenolic compounds from cell vacuoles after cell membranes have lost their semipermeable properties. Insoluble Si accumulation, as determined by energy dispersive X-ray microanalysis using a scanning electron microscope, did not directly correspond to the initial appearance of autofluorescence in HR-dead cells, but occurred gradually after death and in apparent relation to increasing autofluorescence intensity. An hypothesis is stated that phenolics are released by decompartmentalization following cell death and accumulate in the dead cells wall area causing autofluorescence, and that these phenolics form insoluble complexes with Si moving apoplastically in epidermal wall areas due to passive, transpiration stream transport.

Hidden robbers: The role of fungal haustoria in parasitism of plants

Biotrophic fungi have developed a range of “life styles” in their relationship with plants from the mutualistic to the parasitic. Vesicular-arbuscular mycorrhizal fungi form mutualistic relationships with the roots of their plant hosts, in which the fungus obtains sugars from the plant and provides phosphates and other minerals in return. At the other extreme, powdery mildew and rust fungi form an obligately parasitic relationship in which the host plant becomes a source for sugars, amino acids, and other nutrients. These parasites develop a specialized organ, the haustorium (Fig. 1) within plant cells for transfer of nutrients from host cell to fungal thallus. The haustorium is assumed to have a key role in the ability of these parasites to compete with the developing plant for photoassimilates and other nutrients but basic questions remain regarding the function of the haustorium. These include: What are the major nutrients transported? What mechanisms are involved in the transport? How do individual components of the haustorium–host cell interface contribute to nutrient flow? And overall, how does haustorial function relate to the biotrophic relationship between host and parasite? The paper by Voegele et al. (1) in this issue of PNAS provides an important advance by characterizing a sugar transporter located at the haustorium–host interface.

Isolation and expression of a host response gene family encoding thaumatin-like proteins in incompatible oat-stem rust fungus interactions.

Four cDNA clones (corresponding to tlp-1, -2, -3, and -4 genes) encoding thaumatin-like (TL), pathogenesis-related proteins were isolated from oat (Avena sativa) infected by an incompatible isolate Pga-1H of the oat stem rust fungus (Puccinia graminis f. sp. avenae). All four cDNA clones contained an open reading frame predicted to encode a 169-amino acid polypeptide with a signal peptide of 21 amino acids at the N-terminus, suggesting that these proteins are transported through a secretory pathway. The amino acid sequences revealed high homology among the four cDNA clones, 80 to 99% identity and 86 to 100% similarity. The tlp genes and several TL protein genes of certain cereals are clustered into a small group that is phylogenetically separate from the major group of TL protein genes of several plant species. In plants infected with the incompatible isolate Pga-1H, or an inappropriate isolate Pgt-8D of P. graminis f. sp. tritici, high levels of tlp gene transcripts accumulated at 42 to 48 h AI and thereafter when hypersensitive host cell death occurred and hyphal growth was inhibited, whereas in plants infected with a compatible isolate Pga-6A, relatively lower amounts of transcripts were detected. Overall, transcript levels were higher with tlp-1 than with the three other genes. Spray with a light mineral oil used as a spore carrier induced transient expression of tlp-1, -2, and -3 genes at 16 to 30 h AI which obscured the initial induction of the tlp genes in response to infection by the pathogens. In contrast, tlp-4 was induced very little by oil spray, so that induction was clearly observed in response to either compatible, incompatible, or inappropriate isolates at 24 to 30 h AI. Wounding leaves by either slicing or puncturing them strongly induced tlp-1 and tlp-3, moderately induced tlp-2, but had no effect on tlp-4. Taken together, the results showed that tlp genes displayed differential responses to oil spray, mechanical wounding, and pathogen infection and that the expression of tlp genes, especially tlp-1, in oat is associated with resistance reactions in response to infection by incompatible and inappropriate isolates of the stem rust fungi.

Morphology, Life Cycle Biology, and DNA Sequence Analysis of Rust Fungi on Garlic and Chives from California

ABSTRACT In the late 1990s, commercial garlic fields in California (CA) were devastated by an outbreak of rust caused by Puccinia allii. We compared collections of the pathogen from garlic (Allium sativum) and chives (A. schoenoprasum) in central CA and Oregon (OR) to collections from garlic and leek (A. porrum and A. ampeloprasum) in the Middle East. Teliospores from the CA and OR collections were smaller in length, width, and projected cross-sectional area compared with collections from the Middle East. CA and OR collections had a shortened life cycle, in which pycnia and aecia were not formed. Germinating teliospores produced a two-celled promycelium, resulting in two basidiospores, each initially with two nuclei, indicating that this rust was homothallic. In addition, the morphology of the substomatal vesicles was different between the CA-OR (fusiform) and the Middle Eastern (bulbous) collections. DNA sequence analysis of the nuclear ribosomal internal transcribed spacer region showed that the CA and OR rust collections formed a well-supported cluster distinct from the Middle Eastern and European samples. These results suggest that the rust on garlic and chives in CA and OR is a different species than the rust fungus on garlic and leek in the Middle East.

Incompatibility conditioned by the Mla gene in powdery mildew of barley: timing of the effect of cordycepin on hypersensitive cell death

The effect of cordycepin, an inhibitor of mRNA synthesis, on the hypersensitive host cell death response (HR), was tested with epidermal tissues from barley coleoptiles containing the Mla gene for resistance, inoculated with incompatible race 3 of Erysiphe graminis f.sp. hordei . In untreated control tissues, HR occurred 18–30 h after inoculation in 61–91% of cells with haustoria. The percentage of HR was strongly reduced by cordycepin at 10–31 μ m but only if applied 4 h after inoculation or before; not if applied 8 h or later. Effective cordycepin treatments had no inhibitory effect on the development of germ tubes or appressoria and usually increased the development of haustoria and hyphae. The results indicate that mRNAs required for the HR are synthesized in the host within the first 0–8 h after inoculation and that mRNA synthesis is not required thereafter. The timing of protein synthesis required for the HR in the host could not be determined because the inhibitors of protein synthesis used, blasticidin S and cycloheximide, inhibited parasite development at concentrations required to inhibit the HR.

Permeability to nonelectrolytes as an indicator of change in membrane lipid structure in Hordeum vulgare cells infected by Erysiphe graminis f. sp. hordei

The passive permeability to water and nonelectrolytes of epidermal cells from partially dissected coleoptiles of Hordeum vulagare was investigated in uninfected tissue and in tissue infected with a compatible race of the powdery mildew fungus, Erysiphe graminis f. sp. hordei . Permeability constants were determined plasmometrically for water and for the increasingly nonpolar electrolytes: urea, methyl urea, and ethyl urea. Measurements were taken 48 h after inoculation, usually in heavily infected tissues which exhibited concave plasmolysis. Permeability of cells to water or urea was unchanged by infection. Permeability to methyl urea was reduced significantly in the cultivar Montcalm, but not consistently in Atlas or AlgS. Permeability to ethyl urea in AlgS was reduced by half by infection. The reduced permeability occurred not only in haustoriumcontaining cells. but throughout heavily infected tissue. If tissues were left overnight on plasmolyzing solution of α-methyl glucose, uninfected tissues deplasmolyzed and regained turgor whereas infected tissue did not, suggesting that the α-methyl glucose entered healthy, but not infected cells. The results indicate that infection altered the plasmalemma of host cells, reducing passive permeability to α-methyl glucose and the more lipophilic of the nonelectrolytes tested.

Incompatibility conditioned by the Mla gene in powdery mildew of barley: change in permeability to non-electrolytes☆☆☆

Abstract Passive permeability of host cells to the non-electrolytes, methyl and ethyl urea, was measured plasmometrically in epidermal tissue from barley ( Hordeum vulgare ) inoculated with Erysiphe graminis f.sp. hordei . Incompatibility, conditioned by the Mla gene for resistance, was expressed by hypersensitive collapse of haustorium-containing host cells, 1–2 h after cytoplasmic streaming stopped. Measurements were made about 20 h after inoculation, either before or after the halt in streaming. Before streaming stopped, permeability of haustorium-containing cells to methyl and ethyl urea was reduced about 30%. Permeability to methyl urea was reduced also in neighboring cells without haustoria. Cells in tissues with decreased permeability exhibited high frequencies of concave plasmolysis, indicative of increased adhesion between the plasmalemma and the cell wall. After cytoplasmic streaming stopped, but prior to cell collapse, permeability to ethyl urea was reduced to 50% of values for non-inoculated tissues. The results indicate that structural changes which reduce permeability to non-electrolytes occur in the lipid bilayer of the host plasmalemma and that these changes that may be related to reduced membrane fluidity. These changes are similar in incompatible and compatible host-parasite combinations but occur earlier and to a greater degree with incompatibility.