Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Endomembrane system wikipedia , lookup
Tissue engineering wikipedia , lookup
Extracellular matrix wikipedia , lookup
Cell growth wikipedia , lookup
Cellular differentiation wikipedia , lookup
Cell encapsulation wikipedia , lookup
Cell culture wikipedia , lookup
Organ-on-a-chip wikipedia , lookup
Cytokinesis wikipedia , lookup
Cytoplasmic streaming wikipedia , lookup
The Plant Journal (2003) 33, 775–792 GFP-tagging of cell components reveals the dynamics of subcellular re-organization in response to infection of Arabidopsis by oomycete pathogens Daigo Takemoto, David A. Jones and Adrienne R. Hardham Plant Cell Biology Group, Research School of Biological Sciences, Australian National University, PO Box 475 Canberra, ACT 2601, Australia Received 20 September 2002; revised 21 November 2002; accepted 2 December 2002. For correspondence (fax þ61 2 6125 4331; e-mail [email protected]). Summary Cytoplasmic aggregation, the rapid translocation of cytoplasm and subcellular components to the site of pathogen penetration, is one of the earliest reactions of plant cells against attack by microorganisms. We have investigated cytoplasmic aggregation during Arabidopsis–oomycete interactions. Infection by nonpathogenic Phytophthora sojae was prevented in the plant epidermal cell layer, whereas Peronospora parasitica isolates Cala2 (avirulent) and Noks1 (virulent) could both penetrate into the mesophyll cell layer. Epidermal cell responses to penetration by these oomycetes were examined cytologically with a range of transgenic Arabidopsis plants expressing Green Fluorescent Protein (GFP)-tagged cell components. These included plants containing GFP-TUA6 for visualizing microtubules, GFP-hTalin for actin microfilaments, GFP-tm-KKXX for endoplasmic reticulum (ER), and STtmd-GFP for the Golgi apparatus. In all interactions, actin microfilaments were actively re-arranged and formed large bundles in cytoplasmic strands focused on the penetration site. Aggregation of ER membrane and accumulation of Golgi bodies at the infection site were observed, suggesting that production and secretion of plant materials were activated around the penetration site. Microtubules did not become focused on the penetration site. No difference was evident between the responses of epidermal cells in the non-host, incompatible and compatible interactions. This result indicates that the induction of cytoplasmic aggregation in Arabidopsis epidermal cells was neither suppressed by the virulent strain of Peronospora, nor effective in stopping infection. Keywords: Arabidopsis thaliana, oomycetes, actin microfilaments, microtubules, Golgi apparatus, endoplasmic reticulum. Introduction Recent use of Green Fluorescent Protein (GFP), fused to target proteins, has enabled the study of protein localization in living plant cells and the labeling of selected organelles or cell components. There have been many reports of the successful visualization of a variety of cell structures, including microtubules (Marc et al., 1998; Ueda et al., 1999), microfilaments (Klahre et al., 2000; Kost et al., 1998), endoplasmic reticulum (ER) (Benghezal et al., 2000; Haseloff et al., 1997), the Golgi apparatus (Boevink et al., 1998; Nebenführ et al., 1999), mitochondria (Köhler et al., 1997; Niwa et al., 1999), vacuoles (Di Sansebastiano et al., 1998; Mitsuhashi et al., 2000), and peroxisomes (Cutler et al., 2000; Mano et al., 1999) using GFP-tagging. This approach ß 2003 Blackwell Publishing Ltd is an extremely powerful tool to monitor cell dynamics over time in living cells during particular processes such as cell division, vesicle and organelle trafficking, and in response to various kinds of stimuli including temperature, light, wounding, and pathogen attack. Plant cell responses involved in the early stages of attack by microorganisms have been of interest to plant pathologists for many years. The existence of cytoplasmic aggregation was first reported by Tomiyama (1956) in the potato– Phytophthora system. This was defined as the rapid translocation of cytoplasm and subcellular components to the site of pathogen penetration of the cell. Similar plant responses to pathogens were observed in cowpea–Uro775 776 Daigo Takemoto et al. myces (Heath and Heath, 1971), barley–Erysiphe (Kunoh et al., 1985), parsley–Phytophthora (Gross et al., 1993), flax– Melampsora (Kobayashi et al., 1994), onion–Magnaporthe (Xu et al., 1998), and onion–Botrytis (McLusky et al., 1999) interactions. This indicates that cytoplasmic aggregation is a common phenomenon in the early response of plant cells to pathogens in both non-host and race-specific interactions. Pharmacological studies in the early 1980s first suggested the possibility that cytoplasmic aggregation during the defense response was dependent on cytoskeletal reorganization in plant cells (Hazen and Bushnell, 1983; Tomiyama et al., 1982). Cytochalasins, inhibitors of the polymerization of actin, inhibited or delayed resistance reactions such as cell death (Hazen and Bushnell, 1983; Škalamera and Heath, 1998; Takemoto et al., 1999; Tomiyama et al., 1982), callose deposition (Kobayashi et al., 1997a; Škalamera and Heath, 1996), and pathogenesis-related (PR) protein expression (Takemoto et al., 1999), as well as cytoplasmic aggregation. Moreover, treatment with cytochalasins and microtubule polymerization and depolymerization inhibitors permitted the non-pathogen Erysiphe pisi to penetrate barley coleoptile cells and form haustoria and secondary hyphae (Kobayashi et al., 1997a). These results indicated that the cytoskeleton may play an important role in plant resistance; however, details of the precise contribution of cytoskeletal elements have not yet been elucidated. As is generally the case in inhibitor studies, it is difficult to determine what effects of the treatments apply specifically to the defense response and what may be due to interference in other fundamental functions played by the cytoskeleton in plant cell growth and development (see reviews by Nick, 1999; Staiger, 2000). Over the last decade, a large amount of information on plant–pathogen interactions has been obtained from studies of the infection of Arabidopsis by pathogens such as Peronospora parasitica and Pseudomonas syringae. More than 10 resistance genes (R-genes) have been characterized at a molecular level (see review by Holub, 2001). Other components involved in the expression of resistance have also been identified, such as PR proteins (e.g. PR-1 to -5), enzymes required for phytoalexin biosynthesis (e.g. PAD3), signal transduction factors (e.g. NPR1, EDS1, and PAD4), and possible rate-limiting factors of defense (e.g. RAR1 and STG1) (Austin et al., 2002; Muskett et al., 2002; Parker et al., 2000; Ryals et al., 1996; Zhou et al., 1999). However, cytological studies of Arabidopsis–pathogen interactions are relatively limited. There are some reports of the aggregation of plant material at the site of pathogen penetration in Arabidopsis. Koch and Slusarenko (1990) described the aggregation of plant material around the penetration hyphae of P. parasitica in both compatible and incompatible interactions. Parker et al. (1993) observed more callose accumulation in the incom- patible interaction between Arabidopsis and P. parasitica than in the compatible interaction. Donofrio and Delaney (2001) investigated callose deposition in the compatible interaction in detail and found that extrahaustorial callose was present on nearly half of the haustoria examined. In the newly established Arabidopsis–Phytophthora porri system, the deposition of plant material and callose was described as a specific response in the incompatible interaction (Roetschi et al., 2001). However, as yet, there are no reports that describe cytoskeletal or endomembrane re-arrangement within Arabidopsis cells in response to pathogen attack. In the present study, we have investigated the response of epidermal cells in cotyledons of A. thaliana to attack by three oomycete pathogens. Cytoplasmic re-organization in response to penetration by P. sojae (non-pathogenic) and P. parasitica isolates Cala2 (avirulent) and Noks1 (virulent) was examined in transgenic Arabidopsis plants in which selected cell components were tagged with GFP. Details of the re-arrangement of cytoskeletal elements and components of the endomembrane system were revealed. Our results indicate that cytoplasmic aggregation in Arabidopsis epidermal cells is not sufficient to stop P. parasitica ingress in the compatible or incompatible interaction. Possible roles of cytoplasmic aggregation and cytoskeletal rearrangement in plant–pathogen interactions are discussed. Results The interaction between Arabidopsis and Phytophthora sojae Phytophthora sojae Kaufmann et Gerdemann (syn. P. megasperma f. sp. glycinea Kuan et Erwin) is the causal agent of stem and root rot of soybean. Phytophthora sojae has only a limited host range and it is known that A. thaliana is not a host for P. sojae (Kamoun, 2001). The interaction between Arabidopsis and P. sojae was examined using light microscopy, after staining with lactophenol trypan blue of inoculated cotyledons, sampled from 6 h to 7 days after inoculation. There were few zoospores moving on the cotyledon surface, 6 h after inoculation, and most had already encysted and germinated. At this point of time, a variety of infection stages of P. sojae were present, from the initiation of cyst germination to the formation of an appressorium-like swelling. One day after inoculation, most cysts had germinated and developed appressorium-like structures which formed preferentially over anticlinal walls between epidermal cells (Figure 1a). Often, germ tubes grew across the surface of the cotyledon and occasionally formed more than one appressorium-like swelling (Figure 1b). In rare instances, P. sojae grew through stomatal openings or penetrated epidermal cells directly through the outer ß Blackwell Publishing Ltd, The Plant Journal, (2003), 33, 775–792 Arabidopsis response to oomycete pathogens epidermal wall (data not shown). Occasionally, finger-like structures that we interpret as developing haustoria formed under the appressorium-like swelling (Figure 1b). In most cases, P. sojae hyphae were unable to reach the plant mesophyll cell layer. Plant cells around the penetration site were generally not stained with trypan blue (staining of plant cells with trypan blue is indicative of cell death), although a few plant cells attacked by multiple P. sojae germlings did undergo cell death (data not shown). In no case was P. sojae found to colonize the mesophyll layer (Figure 1c). Morphological comparisons of incompatible and compatible interactions between Arabidopsis thaliana and Peronospora parasitica The extent of Peronospora development in incompatible and compatible interactions with Arabidopsis was examined. To monitor pathogen growth and plant cell death, whole cotyledons of A. thaliana ecotype Columbia (Col-0) were stained with lactophenol trypan blue 6 h to 7 days after inoculation with P. parasitica isolates. Col-0 possesses the RPP2 resistance gene which confers resistance to P. parasitica isolate Cala2 but not to P. parasitica isolate Noks1 (Holub et al., 1994). Thus, the interaction of Col-0 and Cala2 is incompatible and that between Col-0 and Noks1 is compatible. Six hours after inoculation, most of the Cala2 or Noks1 conidia had germinated and penetrated the underlying epidermis, although it should be noted that ungerminated conidia tended to be lost during processing. Both Cala2 and Noks1 preferentially penetrated between the anticlinal walls of adjacent epidermal cells, and at 6 h after inoculation, both had started forming haustoria in the epidermal cells (Figure 1d,j). By 12 h, mature haustoria had formed below the penetration site, and for both isolates, about 10% of the haustoria in epidermal cells were encapsulated by plant cellular material (Figure 1l). A difference between the incompatible and compatible interactions became evident 1 day after inoculation. In the Col-0–Cala2 interaction, densely-stained cells containing mature or developing haustoria were observed in the mesophyll layer (Figure 1e,f). Cala2 mycelium growth usually terminated within a cluster of dead mesophyll cells, but hyphae were seen occasionally beyond such a cluster (Figure 1g). Blue-stained epidermal cells were occasionally found next to the penetration site (Figure 1i) but did not prevent growth of hyphae. Four days after inoculation, there were many clusters of two to five dead mesophyll cells in the Col-0–Cala2 interaction and some with more than 10 dead cells (Figure 1h). Inoculated seedlings continued to grow normally, and conidiophore formation and macroscopically visible symptoms of infection were never observed for the Col-0–Cala2 interaction. ß Blackwell Publishing Ltd, The Plant Journal, (2003), 33, 775–792 777 At 1 day after inoculation in the compatible interaction between Col-0 and Noks1, a number of large haustoria had formed within the mesophyll layer in most infection sites, and further mycelial growth and subsequent formation of additional developing haustoria were observed (Figure 1k). Blue-stained epidermal cells occasionally occurred at the penetration site as seen for the Col-0–Cala2 interaction. Two to three days after inoculation, extensive growth of intercellular mycelia and the formation of large numbers of haustoria were evident (Figure 1m). Mesophyll cell death, indicated by dense staining, was observed in only a few instances where there was massive infection. Up to this stage, there were no macroscopically visible symptoms on the inoculated cotyledon. Four days after inoculation, oogonia and antheridia appeared (Figure 1n). Conidiophores emerged 4 to 5 days after inoculation (Figure 1o) and a lawn of conidiophores was macroscopically obvious by 6 days on most of the inoculated cotyledons. The seedlings became etiolated and finally collapsed by about 10 days. In experiments in which transgenic Col-0 plants expressing GFP-TUA6, STtmd-GFP, GFP-tm-KKXX, or GFP-hTalin were inoculated with P. parasitica, no differences in the response of wild-type and transgenic plants to Noks1 or Cala2 were detected (data not shown). Quantitative comparison of the incompatible and compatible interactions between Arabidopsis and Peronospora parasitica After lactophenol trypan blue staining, the numbers of haustoria formed in mesophyll and epidermal cells and the number of dead plant cells per infection site were scored for a large number of penetration sites from 6 to 48 h after inoculation, in order to compare the incompatible (Col-0–Cala2) and the compatible (Col-0–Noks1) interactions. In both interactions, more than 99% of the conidia that remained on the cotyledon after processing produced hyphae that penetrated between the anticlinal walls of two epidermal cells. Up to 24 h after inoculation, the numbers of developing and mature haustoria per infection site were almost the same for both interactions. During this period, the number of haustoria per infection site did not increase dramatically, although many matured (compare the data for 6 h with that for 24 h after inoculation in Figure 2a). However, from 24 to 48 h after inoculation, Noks1 showed massive production of haustoria while Cala2 produced only a small number of additional haustoria from a few mycelia that were not encased by dead plant cells (Figure 2a). This dramatic difference was due mainly to haustoria production by Noks1 in mesophyll cells because the numbers of haustoria formed in epidermal cells were almost the same for both interactions over the time examined (Figure 2b). At 6 and 12 h after inoculation, almost no plant cell death was observed in either interaction. Differences in 778 Daigo Takemoto et al. ß Blackwell Publishing Ltd, The Plant Journal, (2003), 33, 775–792 Arabidopsis response to oomycete pathogens 779 the frequency of plant cell death between the two interactions became evident 24 h after inoculation (0.29 0.04 dead cells per infection site for the incompatible interaction, 0.04 0.02 for the compatible). This difference became more pronounced at 48 h (0.95 0.08 for incompatible, 0.13 0.04 for compatible) (Figure 2c). However, as was the case for haustorium formation, the frequency of cell death in the epidermis was the same in both interactions during the time period studied (0.06 0.02 for both incompatible and compatible) (Figure 2c). Microtubule distribution in Arabidopsis epidermal cells during infection by oomycete pathogens Transgenic Arabidopsis plants expressing GFP-tagged a-tubulin (GFP-TUA6) (Ueda et al., 1999) were used to visualize microtubules in living cotyledon cells. In uninoculated plants, the microtubules formed cortical arrays of gently curving, brightly fluorescent strands underlying the outer epidermal cell wall (Figure 3a). Divergence of bright strands into fine strands of lower fluorescence intensity suggests that most of the fluorescent strands represent bundles of microtubules. Occasionally, weakly fluorescent Figure 1. Cytological characterization of three Arabidopsis–oomycete interactions. Arabidopsis thaliana ecotype Columbia (Col-0) was inoculated with Phytophthora sojae (non-pathogen), Peronospora parasitica isolate Cala2 (avirulent), or isolate Noks1 (virulent). Inoculated cotyledons were stained with lactophenol trypan blue. cy, cyst; c, conidium; as, appressorium-like swelling; h, haustorium; ps, penetration site; dh, developing haustoria; eh, encapsulated haustorium, an, antheridium; o, oogonium; cp, conidiophore. Bar ¼ 25 mm (a, b, f, l) and 50 mm (c–e, g–k, m–o). (a–c) Non-host interaction of Col-0 with P. sojae. (a) A germinated cyst with appressorium-like swelling on the surface of cotyledon. The micrograph was taken 6 h post-inoculation (hpi). (b) One day post-inoculation (dpi). The appressorium-like swellings were preferentially formed over the anticlinal walls of plant epidermal cells. A finger-like structure is indicated by the arrowhead. (c) Germ tubes growing across the surface of the cotyledon. Insert at the top right is the region indicated by the square in the main panel but focused within the epidermal cell. Note that P. sojae does not penetrate into the mesophyll cell layer 5 dpi. (d–i) Incompatible interaction of Col-0 with P. parasitica isolate Cala2. (d) Initial stage of the infection by Cala2. Cala2 preferentially penetrates between the anticlinal walls of adjacent epidermal cells and forms some haustoria 6 hpi. (e) Mesophyll cell death beneath the penetration site. Image at the top right is the same penetration site as in the main panel but focused on the mesophyll cell layer. Note that the epidermal cells in direct contact with Cala2 (indicated by arrowheads) do not show cell death 1 dpi. (f) Haustoria formation by Cala2 in epidermal cells. Arrowheads indicate dead mesophyll cells 1 dpi. (g) Clusters of dead mesophyll cells. Arrowheads indicate mycelia growing beyond the cluster of dead cells in the mesophyll layer 3 dpi. (h) Large cluster of dead cells in the mesophyll cell layer 4 dpi. (i) Epidermal cell death adjacent to the penetration site (indicated by black arrowheads). White arrowhead indicates the dead mesophyll cell 1 dpi. (j–o) Compatible interaction of Col-0 with P. parasitica isolate Noks1. (j) Initial stage of the infection by Noks1. Note that Noks1 initially showed a similar pattern of infection as Cala2 in (d) 6 hpi. (k) Haustorial formation and further infection of Noks1 without plant cell death 1 dpi. (l) Fully developed and encapsulated haustoria formed in epidermal cell 1 dpi. (m) Heavy colonization by Noks1 with numerous haustoria 3 dpi. (n) Oogonia and antheridia formation 4 dpi. (o) Conidiophore formation 5 dpi. ß Blackwell Publishing Ltd, The Plant Journal, (2003), 33, 775–792 Figure 2. Time course of haustoria formation and induction of plant cell death during the infection of avirulent (Cala2) and virulent (Noks1) Peronospora parasitica. (a) Haustoria formation. (b) Haustoria formation in epidermal cells. (c) Induction of plant cell death. Haustoria or plant cell death were scored by observation of inoculated plants after clearing and staining in lactophenol trypan blue. The numbers of infection sites assessed are indicated above each column. cytoplasmic strands were seen running beneath the cortical microtubule array in these cells (Figure 3a). Radial arrays of microtubules were seen in stomatal guard cells (Figure 3a). As in wild-type plants, 6 h after inoculation with zoospores of P. sojae, many appressorium-like swellings had 780 Daigo Takemoto et al. Figure 3. Distribution of microtubules in plant epidermal cells after penetration by oomycete pathogens. Images were collected 6 h after inoculation. Penetration sites and cytoplasmic strands are indicated by asterisks and closed arrowheads, respectively. The junctions of epidermal cells are outlined by dashed lines. Except for (c), images of Green Fluorescent Protein (GFP) fluorescence are projections of optical sections taken at 1-mm intervals from the outer epidermal wall through to immediately above the cortical cytoplasm adjacent to the inner periclinal wall of the epidermal cell. Bar ¼ 10 mm. (a) Tubulin network visualized by GFP-TUA6 in an uninoculated plant. (b,c) GFP-TUA6 plant inoculated with the non-pathogen Phytophthora sojae. (b) Diffuse fluorescence has accumulated around the penetration site. Weakly fluorescent cytoplasmic strands running through the cell focus on the penetration site (b, c). Image in (c) shows a single optical section taken at the centre of the epidermal cell. (d–f) GFP-TUA6 plant inoculated with the avirulent pathogen Peronospora parasitica isolate Cala2. Microtubules do not focus on the penetration site but diffuse fluorescence is seen around the penetration site and in cytoplasmic strands (closed arrowheads) directed toward this region. A tendency for the microtubules to form a circumferential arrangement around the penetration site is evident to varying degrees (most clearly seen in (f)). (g–i) GFP-TUA6 plant inoculated with the virulent pathogen P. parasitica isolate Noks1. The pattern of fluorescence of GFP-TUA6 is similar to that observed after inoculation with Cala2. Diffuse fluorescence is seen around the penetration site and in cytoplasmic strands (closed arrowheads) directed toward this region. Microtubules are not focused on the penetration site but in some cases form an approximately circumferential arrangement in cells surrounding the penetration site (e.g. (h)). formed on the cotyledons of the GFP-TUA6 plants, and hyphae had attempted to grow between the anticlinal walls of the epidermal cells. Observation of GFP-TUA6 fluorescence revealed enhanced but diffuse fluorescence around the penetration site (Figure 3b). Similar images were seen following inoculation of the GFP-TUA6 plants with avirulent (Figure 3d–f) and virulent (Figure 3g–i) isolates of P. parasitica. Fluorescent microtubule bundles could be seen to ß Blackwell Publishing Ltd, The Plant Journal, (2003), 33, 775–792 Arabidopsis response to oomycete pathogens run through the region of diffuse fluorescence. This type of diffuse fluorescence was not observed in wild-type plants and is not due to autofluorescence of material deposited around the penetration site (data not shown). In many cases, the density and arrangement of microtubule bundles in the cortical arrays near the penetration site did not appear to be noticeably different from that seen in uninoculated plants. There was no evidence of microtubule bundles being specifically focused on the penetration site (Figure 3b,e,f,h,i). However, in some cells, the microtubules appeared to be aligned circumferentially around the penetration site (Figure 3d,f,h). This was a global arrangement that encompassed a number of epidermal cells in the vicinity of the penetration site and was seen in all three interactions. In addition, in cells adjacent to the penetration site, greater numbers of broad, diffusely fluorescent cytoplasmic strands were observed traversing the central vacuole of the cells (Figure 3c), and these did show a tendency to focus on the infection site (Figure 3b,e,f,h). These patterns of GFP-TUA6 fluorescence persisted in all three Arabidopsis–oomycete interactions throughout the duration of observations (up to 36 h after inoculation), except that GFP fluorescence was not seen in epidermal cells that had died (data not shown). Re-organization of microfilaments in Arabidopsis epidermal cells during oomycete infection The actin microfilament network was observed in living epidermal cells of transgenic Arabidopsis plants expressing GFP conjugated to the actin-binding domain (McCann and Craig, 1997) of human talin (GFP-hTalin). The brightness of the fluorescence of the GFP-hTalin varied in the 14 independent transgenic lines that were isolated. In some lines, presumably with high levels of transgene expression, the network of fine cortical microfilaments beneath the surface of epidermal cells could be clearly seen in addition to the cytoplasmic microfilament cables in the centre of the cells (Figure 4a). However, these transgenic plants tended to show growth inhibition (e.g. shorter plants and smaller leaves) compared with wild-type plants, and the cotyledons tended to be angled downward rather than being held horizontally. Nevertheless, these plants were still able to grow and set viable seeds (data not shown). In transgenic plants with apparently intermediate levels of GFP-hTalin expression, microfilaments in cytoplasmic strands were still very prominent, but the fine microfilaments in the cell cortex were more weakly fluorescent and difficult to see than in the plants expressing the transgene at a high level (Figure 4b,c). However, as no effects on growth or morphology were detected in these latter plants, one line was selected and used for inoculation with the oomycete pathogens. After their penetration along the anticlinal walls of epidermal cells, similar patterns of actin microfilament distriß Blackwell Publishing Ltd, The Plant Journal, (2003), 33, 775–792 781 bution were observed in the GFP-hTalin plants inoculated with the non-pathogen P. sojae (Figure 4d–f) and the incompatible (Figure 4g–i) and compatible (Figure 4j–l) isolates of P. parasitica. Many relatively fine microfilaments and a number of strongly fluorescent bundles of microfilaments became focused on the penetration site. In many cases, small bundles of microfilaments coalesced to form larger bundles near the penetration site (Figure 4e,i,k, closed arrowheads). Re-organization of the microfilament network was very active around the penetration site (Figure 5). During the 2-min intervals between the images shown in Figure 5, both the more weakly fluorescent fine microfilaments as well as the strongly fluorescent microfilament cables change their position, coalesce, or diverge. Changes in structure of the endoplasmic reticulum network after inoculation with oomycete pathogens The Arabidopsis plants used to determine the distribution of ER after inoculation with the three oomycete pathogens were transformed with a chimaeric gene encoding GFP fused between the signal peptide from Arabidopsis basic chitinase at the N-terminus and the transmembrane domain and short cytosolic tail containing the KKXX motif of the tomato Cf-9 resistance protein (Jones et al., 1994) at the C-terminus (Benghezal et al., 2000). The C-terminal KKXX signal (dilysine motif) confers ER localization to membrane-bound GFP (Benghezal et al., 2000). In GFPtm-KKXX plants, typical reticulate ER was clearly observed in uninoculated plants as previously reported (Benghezal et al., 2000). After inoculation with any of the three oomycete pathogens, intense fluorescence rapidly accumulated around the penetration site (Figure 6). The dynamics of this increase in fluorescence around the invading hypha are illustrated in Figure 6(a) after inoculation with P. sojae. In this example, aggregation of ER is seen to begin before the hypha has penetrated between the epidermal cells. We interpret the strong fluorescence in cells on either side of the penetration site as constituting sheet-like lamellae of ER cisternae. These lamellae were continuous with the tubular ER network that extended throughout the cortical cytoplasm underlying the outer epidermal cell walls. In many instances, the mesh size of the ER network was smaller (about 1–2 mm), closer to the penetration site than it was further away (Figure 6c, open arrowheads; Figure 6b,g,i). The tubular cisternae surrounding the intensely fluorescent ER area at the penetration site sometimes expanded into small lamellar cisternae (e.g. Figure 6f,i). Cytoplasmic strands were invariably brightly fluorescent and were also generally directed to the penetration site (Figure 6b,e,g,i, closed arrowheads). The fluorescence of the cytoplasmic strands tended to be more diffuse than that of the tubular ER, a feature that is likely to be due to movement of the ER 782 Daigo Takemoto et al. ß Blackwell Publishing Ltd, The Plant Journal, (2003), 33, 775–792 Arabidopsis response to oomycete pathogens within the rapidly streaming strands. In a number of instances, the fluorescence of the cortical ER network in cells adjacent to the penetration site was lower in regions distant from the penetration site (Figure 6d, x) than it was close to the penetration site or in cells not immediately adjacent to the infection site (Figure 6d, y). Following inoculation with either Cala2 or Noks1 isolates of P. parasitica, the ER was seen to surround the neck but not the distal regions of haustoria in the epidermal cells (Figure 6j). This lack of fluorescence at the distal end of the haustoria was observed in all of more than 20 haustoria examined. Localization of the Golgi apparatus after inoculation with oomycete pathogens Boevink et al. (1998) reported the use of the transmembrane domain and signal anchor sequences of rat 2,6-sialyl transferase (STtmd) to label the Golgi apparatus in plant cells. In the present study, transgenic Arabidopsis plants expressing STtmd-GFP were used to investigate the distribution of the Golgi apparatus in cells penetrated by oomycete pathogens. As shown in Figure 7, Golgi stacks became preferentially localized around the penetration site in all three Arabidopsis–oomycete interactions. In contrast to the situation for the ER, in which the brightly fluorescent lamellae remained around the penetration site, the Golgi stacks continued to move around the cell and the degree of accumulation at the penetration site changed over time (Figure 7c). Most of the Golgi stacks were transported along cytoplasmic strands running through the centre of the epidermal cells and focusing on the penetration site. Golgi stacks moved towards the penetration site on some strands and away from this site on other strands (data not shown). As for the ER, Golgi stacks surrounded the neck but not the haustorial body of Peronospora haustoria that formed in the epidermal cells (Figure 7f). This lack of Golgi bodies around the distal end of the haustorium was observed in all of more than 20 haustoria examined. Discussion Green Fluorescent Protein has become a valuable tool for investigations of subcellular structures in living cells. Although transgenic plants with various GFP-tagged cell 783 components have been available (see Introduction section), no reports of the use of these plants in studies of cell reorganization in response to pathogenic attack have yet been published. In the present study, we first characterized the non-host, incompatible and compatible interactions of Arabidopsis with oomycete pathogens, and then monitored the re-organization of GFP-tagged cell components during these interactions. Comparison of the Arabidopsis responses to non-host, incompatible and compatible oomycete pathogens All three oomycete pathogens preferentially penetrated the Arabidopsis cotyledons along the anticlinal walls between epidermal cells, as has been commonly observed for the infection of plant roots and shoots by oomycetes (e.g. Hardham, 2001; Koch and Slusarenko, 1990; Stössel et al., 1980). The progress of infection by P. sojae was halted close to the initial penetration site in the epidermis. This inhibition of further growth of P. sojae occurred in the absence of epidermal cell death, indicating that other mechanisms such as the formation of a barrier at the cell wall effectively stopped pathogen ingress. In the Arabidopsis–P. sojae interaction, epidermal cell death was observed only when multiple hyphae were in contact with a cell, although contact with a single hypha of P. infestans or P. porri can cause epidermal cell death in Arabidopsis (Roetschi et al., 2001; Vleeshouwers et al., 2000). Epidermal cell death was rarely observed during the interaction of Col-0 with the incompatible Cala2 and compatible Noks1 isolates of P. parasitica. For the P. parasitica isolates, no difference in the infection process was detected during the first 24 h after inoculation, while the pathogens grew through the epidermal cell layer. Both only occasionally formed haustoria in the epidermal cells. Cala2 and Noks1 grew through the epidermis and into the mesophyll cell layer, and it is within the mesophyll that differences became evident between the incompatible and compatible interactions with Col-0. Growth of Cala2 was, in general, stopped quickly within the mesophyll, with few haustoria being formed and the hyphae becoming surrounded by dead mesophyll cells. Occasionally, Cala2 hyphae were seen beyond a cluster of dead mesophyll cells. This may reflect the inability of Col-0 to completely restrict growth in Figure 4. Distribution of microfilaments in Arabidopsis epidermal cells after penetration by oomycete pathogens. Images were collected 6 h after inoculation. Penetration sites are indicated by asterisks. The junctions of epidermal cells are outlined by dashed lines. Closed arrowheads indicate sites where microfilament bundles merge to form larger cables. Images of Green Fluorescent Protein (GFP) fluorescence are projections of optical sections taken at 1-mm intervals from the outer epidermal wall through to immediately above the cortical cytoplasm adjacent to the inner periclinal wall of the epidermal cell. In all inoculated plants, both large and small bundles of actin microfilaments have become focused on the pathogen penetration site. Bar ¼ 10 mm. (a–c) Uninoculated GFP-hTalin plants showing relatively high (a), intermediate (b), and low (c) levels of transgene expression. Cortical microfilaments are most clearly visualized in (a). (d–f) GFP-hTalin plants inoculated with the non-pathogen Phytophthora sojae. (g–i) GFP-hTalin plants inoculated with the avirulent pathogen Peronospora parasitica isolate Cala2. (j–l) GFP-hTalin plants inoculated with the virulent pathogen P. parasitica isolate Noks1. ß Blackwell Publishing Ltd, The Plant Journal, (2003), 33, 775–792 784 Daigo Takemoto et al. this interaction, although, contrary to a previous report (Holub et al., 1994), no Peronospora sporulation was observed under our conditions. Even during the interaction of Cala2 with Oystese-1 (Oy-1), an ecotype of Arabidopsis which is highly resistant as it has resistance genes at two different loci (RPP2 and RPP3) conferring resistance to Cala2 (Holub et al., 1994), no epidermal cell death was observed (D. Takemoto, unpublished observations). Our results are consistent with other observations (Botella et al., 1998; Parker et al., 1993), which indicate that while cell death may occur in epidermal cells in some interactions (Koch and Slusarenko, 1990), in general, R-gene resistance of Arabidopsis to P. parasitica is associated with cell death in the mesophyll cell layer rather than the epidermal layer (Figure 8). Reduced image quality in the mesophyll layer restricted observations of GFP fluorescence in the present study to the epidermis. Observations of the GFP-tagged Arabidopsis plants using confocal microscopy clearly revealed cytoplasmic aggregation at the penetration site within the epidermal cells during interactions with all three oomycete pathogens. In other studies, more extensive cytoplasmic aggregation (e.g. Kobayashi et al., 1992) has been observed in resistant than in susceptible plants, but in our study there were no obvious differences between the non-host, incompatible and compatible interactions with respect to the aggregation of GFP-tagged cell components. Whether or not cytoplasmic aggregation was responsible for effectively stopping the infection by P. sojae, despite failing to prevent the ingress of either P. parasitica isolate, remains to be determined. Re-arrangement of the plant cytoskeleton during pathogen infection Figure 5. Sequential images of the microfilament network around a Phytophthora sojae penetration site. Images were collected from 6 h after inoculation at 2-min intervals. Top left panel shows the differential interference contrast (DIC) image for the same area as the other images. The germ tube and appressorium-like swelling are outlined by a dashed black line. The junctions of epidermal cells are outlined by dashed white lines. Arrowheads indicate microfilament strands that have formed or increased in brightness during the time since the previous image. The microfilament network was re-arranged actively around the penetration site. Bar ¼ 10 mm. In all three Arabidopsis–oomycete interactions, actin microfilaments were re-arranged to form large bundles focused on the penetration site. Microfilament bundles from different parts of the cell merged and supported active cytoplasmic streaming directed towards the infection site. This rearrangement of actin microfilaments and the associated formation of cytoplasmic strands focused on the infection site is a consistent feature of the plant response in almost all pathosystems studied to date, including barley–Erysiphe (Baluška et al., 1995; Kobayashi et al., 1992), flax–Melampsora (Kobayashi et al., 1994), cowpea–Uromyces (Škalamera et al., 1998), onion–Magnaporthe (Xu et al., 1998), and onion–Botrytis (McLusky et al., 1999). In some cases, microfilament accumulation was greater in non-host or resistant plants than it was in susceptible plants (e.g. Kobayashi et al., 1992, 1994), although in the cowpea–Uromyces system, microfilament cables were formed around the infection site in the susceptible host more frequently than in the resistant host (Škalamera et al., 1998). Drug ß Blackwell Publishing Ltd, The Plant Journal, (2003), 33, 775–792 Arabidopsis response to oomycete pathogens studies have also shown that de-polymerization of the actin cytoskeleton inhibits formation of the cytoplasmic aggregate and reduces the plant’s ability to inhibit pathogen attack (see Introduction). In contrast to the behavior of the actin microfilaments, the network of cortical microtubules did not become focused on the penetration site in any of the three pathogen interactions. In the GFP-TUA6 plants, the most obvious change was the appearance of strong but diffuse fluorescence around the penetration site. Similar fluorescence was not observed in wild-type plants or in any of the other GFPtransformants, indicating that the fluorescence was not due to autofluorescence of material deposited at the penetration site. We consider it likely that the diffuse fluorescence in the GFP-TUA6 plants is due to monomeric or dimeric GFP-tubulin in the cytoplasm. The accumulation of the GFPtubulin subunits at the penetration site could be due to increased tubulin synthesis, or, more likely, to localized microtubule de-polymerization as has been reported previously in the parsley–Phytophthora interaction (Gross et al., 1993) or after elicitor treatment in tobacco cells (Binet et al., 2001). Our observations of diffuse GFP-tubulin fluorescence in the cytoplasm at the penetration site may reflect the retention of cytoplasmic proteins in the living, non-permeabilized cells in contrast to the loss of soluble proteins in cells permeabilized for immunocytochemical labeling. In some Arabidopsis epidermal cells, there was no apparent change in microtubule arrangement, but in others, microtubules seemed to be aligned in a roughly circumferential array in epidermal cells surrounding the penetration site. This type of alignment has not been previously reported as part of the plant response to pathogen attack, although similar global re-arrangements have been observed in response to wounding (Hush et al., 1990) and during leaf formation (Hardham et al., 1980). In both barley–Erysiphe and flax–Melampsora interactions, microtubules become radially arranged and focused on the site of contact with the pathogen (Kobayashi et al., 1992, 1994). We note that in uninoculated plants, many cortical microtubules abut the anticlinal wall junctions at sharp angles, and it is these microtubules that are apparently missing in the cells forming part of the circumferential array. Perhaps the circumferential arrangement we have observed has no significance in itself, but is merely the result of preferential de-polymerization of microtubules abutting the wall adjacent to the penetrating hypha. Thus, while the re-organization of actin microfilaments at the site of pathogen attack is consistent, the behavior of microtubule arrays is quite variable in different plant–pathogen systems. This may be because microtubules do not play a direct role in the response, although a number of studies have shown that microtubule de-polymerization reduces the plant’s resistance (Kobayashi et al., 1997a,b). Alternatively, differences ß Blackwell Publishing Ltd, The Plant Journal, (2003), 33, 775–792 785 in the behavior of microtubule arrays may be influenced by such factors as (i) the species of plant and pathogen involved in the interaction, (ii) whether the pathogen is a necrotroph or biotroph, (iii) whether the pathogen is growing intercellularly or is attempting to penetrate through the plant cell wall, or (iv) whether the system involves cultured cells or intact plant tissues. A number of Arabidopsis-based necrotrophic and biotrophic pathosystems have now been developed, involving true fungi such as Alternaria brassicicola (e.g. Penninckx et al., 1996), Erysiphe spp. (e.g. Adam et al., 1999), Fusarium oxysporum (Mauch-Mani and Slusarenko, 1994), and Botrytis cinerea (e.g. Thomma et al., 1998), as well as the oomycete pathogens P. porri (Roetschi et al., 2001), Albugo candida (Holub et al., 1995), and P. parasitica (Koch and Slusarenko, 1990). Future investigations of the response of the GFP-tagged Arabidopsis plants described in the present study to this range of pathogens may help elucidate the consistent and important features of the cytoskeletal re-organization in these interactions. Re-organization of the endomembrane system during pathogen infection The GFP-tm-KKXX plants revealed dramatic re-organization of the ER at the penetration site after inoculation with the three oomycete pathogens. Strong fluorescence around the penetration site indicated the existence of a dense network of lamellar ER. The lamellar network merged into a condensed reticulum of tubular ER. This latter structure resembled the transitional form of ER between ‘perforated sheets’ and tubular reticulum observed in growing cells (Ridge et al., 1999). The function of the perforated sheets of ER in growing cells is not clear, but may be involved with the synthesis of large quantities of proteins and lipids. A similar function could be envisaged in plant cells responding to pathogen attack because the production of many defense proteins is induced in these cells. A similar reorganization of ER was reported in the compatible interaction of pea and E. pisi, following labeling of the ER by antiHDEL antibodies (Leckie et al., 1995). In this latter study, it was suggested that the accumulation of ER might be involved in exocytosis to generate new membrane and materials to form the haustorial complex. This seems unlikely to be the case in the Arabidopsis–oomycete interactions in the present study because the accumulation of ER membrane at the penetration site occurred in cells with or without haustoria, and in cells containing haustoria, the ER did not encompass the haustorium body. Concentration of ER around the neck of the haustorium has also been observed in the compatible interaction between wheat and Puccinia graminis (Harder et al., 1978). In contrast to the large accumulation of ER around the penetration site, fluorescence of the tubular ER cisternae was weaker on the opposite side of the cell. Cytoplasmic 786 Daigo Takemoto et al. ß Blackwell Publishing Ltd, The Plant Journal, (2003), 33, 775–792 Arabidopsis response to oomycete pathogens strands focused on the penetration site contained ER cisternae, and the accumulation of ER membrane occurred quite rapidly. It is thus likely that much of the accumulated ER membrane was recruited from the existing ER network throughout the cell. Although the reticulate cortical ER usually does not show large-scale translocations and is distinctive in appearance from the ER tubules in streaming cytoplasm (Quader and Schnepf, 1986; Ridge et al., 1999), the ER may re-distribute dramatically in particular circumstances. An example of this is the re-distribution of ER from the cortical cytoplasm into the phragmosome during cell plate assembly (Hepler, 1982; Nebenführ et al., 2000). Inoculation of Col-0 with the three oomycete pathogens led to a concentration of Golgi stacks around the penetration site. In contrast to the stable accumulation of ER, observation of the same penetration site over a period of time revealed that the distribution of Golgi stacks changed continually. Although individual Golgi stacks moved towards or away from the penetration site, the localized accumulation of Golgi stacks indicated that they stopped at least temporarily near the penetration site. Golgi stacks have recently been shown to move along cytoplasmic strands in tobacco leaf cells in an acto-myosin-dependent manner (Boevink et al., 1998; Nebenführ et al., 1999, 2000). The GFP-labeled Golgi stacks displayed stop-and-go movements, oscillating between directed movement and random ‘wiggling’. Stacks moving along the same track often paused at the same position, indicating that there were specific sites that inhibit the movement. From this observation, the authors proposed a ‘recruitment model’ in which Golgi stacks are recruited to active ER export sites (or to the sites for secretion) by localized stop signals that transiently prevent the stacks from leaving the area (Nebenführ et al., 1999). If this model is applied to the Arabidopsis–oomycete interactions, it is possible that the accumulation of ER at the penetration site might be associated with a large quantity of these ‘stop signals’, leading to the preferential accumulation of Golgi stacks. This would favor effective ER-to-Golgi transport and site-specific secretion in response to pathogen penetration. 787 Concluding remarks The aggregation of cytoplasm at the site of pathogen attack is generally believed to be associated with the deposition of material that will generate or strengthen the physical and chemical barriers to pathogen ingress within the plant cell wall. Accumulation of ER and Golgi bodies in the cytoplasmic aggregation is consistent with this role. The function of cytoplasmic aggregation and the cell components responsible for its formation have been investigated mainly in pharmacological studies using inhibitors of cytoskeletal elements. These studies have shown that de-polymerization of either actin microfilaments or microtubules may prevent or delay both non-host and race-specific plant resistance (see Introduction), and that de-polymerization of actin prevents cytoplasmic aggregation. The cytoskeleton is thus thought to play an important part in the mobilization of the plant defense response. Currently, little is known about the factors that induce cytoplasmic aggregation at the infection site. Pathogen attack may generate both physical and chemical signals that can be detected by the underlying plant cell. It has, for example, been shown that local mechanical pressure can trigger the translocation of cytoplasm to the site of stimulation (Gus-Mayer et al., 1998; Kennard and Cleary, 1997), indicating that cytoplasmic aggregation could be induced by the pressure generated by the penetrating hypha. However, in the barley–Erysiphe (Kunoh et al., 1985) and flax– Melampsora (Kobayashi et al., 1994) interactions, it has been observed that cytoplasmic aggregation or cytoskeletal re-organization in the plant cell was induced before pathogen penetration began. In addition, the mitogen-activated protein kinase (mps1) mutant of M. grisea, which is unable to form a penetration peg, still elicits cytoplasmic aggregation on non-host onion cells (Xu et al., 1998). All these data support the idea that a chemical elicitor released by the pathogen is sufficient to induce cytoplasmic aggregation. Responses, such as cytoskeletal re-organization or localized wall modifications in plant cells adjacent to cells in direct contact with the pathogen, could also be triggered by such an elicitor. Figure 6. Re-organization of endoplasmic reticulum (ER) network in Arabidopsis after inoculation by oomycete pathogens. Images were collected 6 h after inoculation. Asterisks indicate the penetration sites. Cytoplasmic strands containing ER are indicated by the closed arrowheads. The junctions of epidermal cells are outlined by dashed lines. Except for (j), images of Green Fluorescent Protein (GFP) fluorescence are projections of optical sections taken at 1-mm intervals from the outer epidermal wall through to immediately above the cortical cytoplasm adjacent to the inner periclinal wall of the epidermal cell. Strong fluorescence around the penetration site indicated the existence of a dense ER network in all inoculated plants. Bar ¼ 10 mm. (a–d) GFP-tm-KKXX plants inoculated with the non-pathogen Phytophthora sojae. (a) Sequential images of the initiation of ER accumulation around the P. sojae penetration site. The germ tube and appressorium-like swelling are outlined by a white dashed line. Images were taken at 5-min intervals. Open arrowheads in (c) indicate regions of the ER network that has a smaller mesh size than in regions distant to the infection site or in uninfected cells. Epidermal cells with (x) or without (y) direct P. sojae contact are marked in (d). (e–g) GFP-tm-KKXX plants inoculated with the avirulent pathogen Peronospora parasitica isolate Cala2. (h–j) GFP-tm-KKXX plants inoculated with the virulent pathogen P. parasitica isolate Noks1. (j) Accumulation of ER membrane around the haustorium. This image is a single optical section focused in the mid-plane of the haustorium. Top left micrograph shows the differential interference contrast (DIC) image of the infection site shown in the main panel. h, haustorium. ß Blackwell Publishing Ltd, The Plant Journal, (2003), 33, 775–792 788 Daigo Takemoto et al. Figure 7. Preferential localization of Golgi stacks around the penetration site of oomycete pathogens. Images were collected 6 h after inoculation. Asterisks indicate the pathogen penetration sites. The junctions of epidermal cells are outlined by dashed lines. Except for (f), images of Green Fluorescent Protein (GFP) fluorescence are projections of optical sections taken at 1-mm intervals from the outer epidermal wall through to immediately above the cortical cytoplasm adjacent to the inner periclinal wall of the epidermal cell. Bar ¼ 10 mm. (a–c) STtmd-GFP plants inoculated with the non-pathogen, Phytophthora sojae. In some cases, the rapid movement of the Golgi stacks during image capture appears as weak trailing signals (indicated by arrowheads). (c) Changes in the accumulation pattern of Golgi stacks at the same penetration site. Images were taken at 15-min intervals. (d–f) STtmd-GFP plants inoculated with the avirulent pathogen Peronospora parasitica isolate Cala2. (f) Accumulation of the Golgi stacks around the haustorium of Cala2. The image is a single optical section focused in the mid-plane of the haustorium. Lower left insert shows the differential interference contrast (DIC) image of the infection site shown in the main panel. h, haustorium. (g–i) STtmd-GFP plants inoculated with the virulent pathogen P. parasitica isolate Noks1. Cytoplasmic aggregation is induced in non-host, incompatible and compatible interactions and is one of the earliest responses to pathogen attack. In many cases, cytoplasmic aggregation and the associated deposition of wall material and toxins may be sufficient to stop infec- tion. Isolation of mutants with enhanced disease susceptibility (e.g. eds, pad, nim1/npri) indicated that even susceptible plants possess a level of basal resistance able to reduce disease severity (Glazebrook et al., 1996). Donofrio and Delaney (2001) investigated callose apposition for ß Blackwell Publishing Ltd, The Plant Journal, (2003), 33, 775–792 Arabidopsis response to oomycete pathogens 789 Figure 8. Schematic diagram summarizing the interactions between Arabidopsis and the three oomycete pathogens. The non-pathogen, Phytophthora sojae, is unable to penetrate through the epidermal cell layer, suggesting that there is effective non-host resistance at this stage. Both avirulent and virulent Peronospora parasitica isolates can form haustoria in the epidermal cells without induction of plant cell death and can reach the mesophyll cell layer. In the incompatible interaction, some haustoria are formed in the mesophyll cells; however, plant cell death occurs in the mesophyll cell layer and further growth of the pathogen is inhibited. In the compatible interaction, numerous haustoria are formed in mesophyll cells. No plant cell death is observed and the virulent pathogen is able to successfully colonize the cotyledon. HR, hypersensitive response. wild type and systemic-acquired resistance (SAR) defective (NahG and nim-1) Arabidopsis following infection by P. parasitica. The NahG and nim-1 plants showed enhanced disease susceptibility and a lower degree of accumulated callose around haustoria compared with wild-type plants. This circumstantial evidence suggests that localized accumulation of plant material may also be involved in basal resistance. Because of the central role of cytoplasmic aggregation in plant defense at various levels, it will be important to determine in greater detail the molecular basis of both its induction and formation. Further studies of the response of GFP-tagged cell components of Arabidopsis to a range of pathogens are likely to provide much new and valuable information in this area. Experimental procedures Plants and growth conditions The transgenic A. thaliana line containing GFP-TUA6 (GFP-tagged alpha-tubulin) (Ueda et al., 1999) was obtained from Dr T. Hashimoto (Nara Institute of Science and Technology, Nara, Japan). Arabidopsis containing STtmd-GFP (sialyltransferase short cytoplasmic tail and single transmembrane domain conjugated with GFP) (Boevink et al., 1998) was obtained from Dr C. Hawes (Oxford Brookes University, Oxford, UK). Generation of Arabidopsis containing GFP-tm-KKXX (GFP with Cf-9 transmembrane domain and cytosolic tail with C-terminal dilysine motif that confers ER localization) was described previously (Benghezal et al., 2000). All transgenic Arabidopsis plants were derived from the Columbia (Col-0) ecotype. Arabidopsis plants were grown at 218C with 16 h of light (50 mmol of photons m2 sec1) per day before pathogen inoculation. To maintain the P. parasitica inoculum, Arabidopsis plants were grown on soil (1 : 1 : 1 mixture of composted soil : sand : peat). For the inoculation assay, seeds of Arabidopsis were surface sterilized and grown on agar plates containing 1 Muraß Blackwell Publishing Ltd, The Plant Journal, (2003), 33, 775–792 shige and Skoog salts (Murashige and Skoog, 1962), 2% (w/v) sucrose, 1 vitamin mix (1000 vitamin mix contained 0.3% (w/ v) thiamine hydrochloride, 0.5% (w/v) nicotinic acid, and 0.05% (w/ v) pyridoxine hydrochloride) and 0.4% (w/v) Agargel (Sigma). Construction of GFP-hTalin fusions and Arabidopsis transformation The ClaI-EcoRI GFP cassette lacking the stop codon was obtained by PCR using pCBJ4 (Benghezal et al., 2000) and two oligonucleotides 50 -ATCGATGTCTAGAGGAGAAGAACTT-30 and 50 -GAATTCTTTGTATAGTTCATCCAT-30 . The EcoRI-BamHI talin actin-binding domain (McCann and Craig, 1997) cassette with 5 Gly-Ala spacer at the N-terminus was amplified by PCR using KIAA1027, containing human talin cDNA (provided by Kazusa DNA Research Institute) and two oligonucleotides, 50 -GAATTCGGCGCTGGGGCAGGTGCCGGAGCTGGAGCGAACTTTGAG-30 and 50 -GGATCCTTAGTGCTCATCTCGAAG-30 . The amplified fragments were cloned into the EcoRV site of Bluescript SK- (Stratagene) using the TA cloning technique. The PCR products were verified by sequencing with ABI Prism BigDye Terminator Cycle Sequencing kits (Perkin-Elmer) according to the manufacturer’s instructions. Both ClaI-EcoRI GFP and EcoRI-BamHI talin actin-binding domain cassettes were ligated into a pUC19 vector containing the cauliflower mosaic virus (CaMV) 35S promoter and the tobacco mosaic virus (TMV) omega leader sequence at the PstI-SalI site to generate 35S-GFP-hTalin/pUC19. The 35S-GFP-hTalin cassette was excised from 35S-GFP-hTalin/pUC19 with HindIII and BamHI. The CaMV 35S terminator sequence was excised from CaMV-GR (Schena et al., 1991) using BamHI and SacI. The 35S-GFP-hTalin cassette and 35S terminator were cloned into HindIII-SacI digested pSLJ7292 (described in http://www.jic.bbsrc.ac.uk/sainsbury-lab/ jonathan-jones/plasmid-list/plasmid.htm) to generate pCBJ-GFPhTalin. Plant transformation was performed by the floral dip method (Clough and Bent, 1998) using Agrobacterium tumefaciens strain AGL-1 (Lazo et al., 1991), containing pCBJ-GFP-hTalin. Transformed seedlings were selected on agar plates (see Plants and growth conditions section) containing 50 mg ml1 kanamycin, and GFP fluorescence monitored under a fluorescence stereo microscope (model MZ FLIII, Leica, Germany). 790 Daigo Takemoto et al. Plant inoculations Inoculation of Arabidopsis plants with the P. parasitica isolates and P. sojae was performed on plants grown for 10–14 days on agar plates. Peronospora parasitica isolates Cala2 and Noks1 were obtained via Dr D. Cahill (Deakin University, Geelong, Australia) from Dr E. Holub (Horticulture Research International, UK) and maintained by weekly subculturing on susceptible recipient plants as described previously (Dangl et al., 1992). Peronospora parasitica conidia (sporangia) were washed from heavily sporulating Arabidopsis leaves with water, collected by centrifugation at 350 g for 5 min and re-suspended in water. For the P. parasitica inoculations, 1–2 ml drops of a suspension of 0.5–1 105 conidia ml1 in water were placed on each cotyledon. Inoculated leaves were kept at 168C with continuous light (50 mmol of photons m2 sec1). Phytophthora sojae (H1169) was maintained on nutrient agar plates (10% (v/v) cleared V8 juice, 0.002% (w/v) b-sitosterol, 0.01% (w/v) CaCO3 and 1.7% (w/v) Bacto agar) at 258C and subcultured 3 days before use. For zoospore production, 3-day-old V8 agar plates with P. sojae mycelium were washed with water every hour over 5 h and left in a final wash overnight at 188C. Released zoospores were collected by centrifugation at 350 g for 5 min and re-suspended in water. For the P. sojae inoculations, 1–2 ml drops of a suspension of 1 104 spores ml1 in water were placed on each cotyledon. Inoculated leaves were kept at 228C with continuous light (50 mmol of photons m2 sec1). Lactophenol trypan blue staining and light microscopy To monitor the progression of oomycete infection and plant cell death, infected leaves were stained as described by Koch and Slusarenko (1990) with minor modifications. Infected leaves were cleared in methanol for more than 24 h and boiled for 3 min in lactophenol trypan blue stain (10 ml H2O, 10 ml of lactic acid, 10 ml of glycerol, 10 g of phenol, and 10 mg of trypan blue). After the leaves had cooled to room temperature for 1 h, the stain was replaced with 1 g ml1 chloral hydrate. Stained leaves were decolorized overnight and viewed using an Axioplan universal microscope (Zeiss, Germany). Confocal laser scanning microscopy Confocal fluorescence and concurrent differential interference contrast (DIC) images were recorded on a TCS SP2 confocal system (Leica, Germany) with a 63 NA 1.2 water-immersion lens. Whole cotyledons were removed from Arabidopsis seedlings 6–36 h after inoculation with one of the three oomycete pathogens and mounted between glass coverslips, with the inoculated surface facing the objective lens. A krypton–argon laser was used as excitation source at 488 nm, and GFP fluorescence was recorded between 495 and 520 nm. Observation of GFP fluorescence was restricted to the epidermal cells because image quality was greatly reduced in the underlying mesophyll layer, due at least in part to the increased depth in the tissue. Except where stated otherwise, images of GFP fluorescence shown in Figures 3–7 are projections of optical sections taken at 1-mm intervals from the outer epidermal wall through to immediately above the cortical cytoplasm, adjacent to the inner periclinal wall of the epidermal cell. The images were stored as TIF files and processed with Adobe Photoshop 4.0 software (Adobe Systems Inc., USA). Acknowledgements We thank Drs Takeshi Hashimoto (Nara Institute of Science and Technology, Japan) for providing GFP-TUA6, Chris Hawes (Oxford Brookes University, UK) for STtmd-GFP plants, and Eric Holub (Horticulture Research International, UK) and David Cahill (Deakin University, Australia) for P. parasitica isolates. We also thank Kazusa DNA Research Institute for providing plasmid KIAA1027, containing human talin cDNA, Professor Jonathan Jones at the Sainsbury Laboratory, John Innes Centre (Norwich, UK) for pSLJ7292, and Daryl Webb for technical assistance with confocal laser microscopy. References Adam, L., Ellwood, S., Wilson, I., Saenz, G., Xiao, S., Oliver, R.P., Turner, J.G. and Somerville, S. (1999) Comparison of Erysiphe cichoracearum and E. cruciferarum and a survey of 360 Arabidopsis thaliana accessions for resistance to these two powdery mildew pathogens. Mol. Plant-Microbe Interact. 12, 1031–1043. Austin, M.J., Muskett, P., Kahn, K., Feys, B.J., Jones, J.D.G. and Parker, J.E. (2002) Regulatory role of SGT1 in early R-genemediated plant defenses. Science, 295, 2077–2080. Baluška, F., Bacigálová, K., Oud, J.L., Hauskrecht, M. and Kubica, Š. (1995) Rapid re-organization of microtubular cytoskeleton accompanies early changes in nuclear ploidy and chromatin structure in postmitotic cells of barley leaves infected with powdery mildew. Protoplasma, 185, 140–151. Benghezal, M., Wasteneys, G.O. and Jones, D.A. (2000) The Cterminal dilysine motif confers endoplasmic reticulum localization to type I membrane proteins in plants. Plant Cell, 12, 1179–1201. Binet, M.-N., Humbert, C., Lecourieux, D., Vantard, M. and Pugin, A. (2001) Disruption of microtubular cytoskeleton induced by cryptogein, an elicitor of hypersensitive response in tobacco cells. Plant Physiol. 125, 564–572. Boevink, P., Oparka, K., Santa Cruz, S., Martin, B., Betteridge, A. and Hawes, C. (1998) Stacks on tracks: the plant Golgi apparatus traffics on an actin/ER network. Plant J. 15, 441–447. Botella, M.A., Parker, J.E., Frost, L.N., Bittner-Eddy, P.D., Beynon, J.L., Daniels, M.J., Holub, E.B. and Jones, J.D.G. (1998) Three genes of the Arabidopsis RPP1 complex resistance locus recognize distinct Peronospora parasitica avirulence determinants. Plant Cell, 10, 1847–1860. Clough, S.J. and Bent, A.F. (1998) Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana. Plant J. 16, 735–743. Cutler, S.R., Ehrhardt, D.W., Griffitts, J.S. and Somerville, C.R. (2000) Random GFP: cDNA fusions enable visualization of subcellular structures in cells of Arabidopsis at a high frequency. Proc. Natl. Acad. Sci. USA, 97, 3718–3723. Dangl, J.L., Holub, E.B., Debener, T., Lehnackers, H., Ritter, C. and Crute, I.R. (1992) Genetic definition of loci involved in Arabidopsis–pathogen interactions. In Method in Arabidopsis Research (Koncz, C., Chua, N.-H. and Schell, J., eds). Singapore: World Scientific Publishing, pp. 393–418. Di Sansebastiano, G.P., Paris, N., Marc-Martin, S. and Neuhaus, J.M. (1998) Specific accumulation of GFP in a non-acidic vacuolar compartment via a C-terminal propeptide-mediated sorting pathway. Plant J. 15, 449–457. Donofrio, N.M. and Delaney, T.P. (2001) Abnormal callose response phenotype and hypersusceptibility to Peronospora parasitica in defense-compromised Arabidopsis nim1–1 and salicylate hydroxylase-expressing plants. Mol. Plant-Microbe Interact. 14, 439–450. Glazebrook, J., Rogers, E.E. and Ausubel, F.M. (1996) Isolation of Arabidopsis mutants with enhanced disease susceptibility by direct screening. Genetics, 143, 973–982. ß Blackwell Publishing Ltd, The Plant Journal, (2003), 33, 775–792 Arabidopsis response to oomycete pathogens Gross, P., Julius, C., Schmelzer, E. and Hahlbrock, K. (1993) Translocation of cytoplasm and nucleus to fungal penetration sites is associated with de-polymerization of microtubules and defence gene activation in infected, cultured parsley cells. EMBO J. 12, 1735–1744. Gus-Mayer, S., Naton, B., Hahlbrock, K. and Schmelzer, E. (1998) Local mechanical stimulation induces components of the pathogen defense response in parsley. Proc. Natl. Acad. Sci. USA, 95, 8398–8403. Harder, D.E., Rohringer, R., Samborski, D.J., Kim, W.K. and Chong, J. (1978) Electron microscopy of susceptible and resistant nearisogenic (sr6/Sr6) lines of wheat infected by Puccinia graminis tritici. Part I. The host–pathogen interface in the compatible (sr6/ P6) interaction. Can. J. Bot. 56, 2955–2966. Hardham, A.R. (2001) The cell biology behind Phytophthora pathogenicity. Aust. Plant Pathol. 30, 91–98. Hardham, A.R., Green, P.B. and Lang, J.M. (1980) Re-organization of cortical microtubules and cellulose deposition during leaf formation in Graptopetalum paraguayense. Planta, 149, 181–195. Haseloff, J., Siemering, K.R., Prasher, D.C. and Hodge, S. (1997) Removal of a cryptic intron and subcellular localization of green fluorescent protein are required to mark transgenic Arabidopsis plants brightly. Proc. Natl. Acad. Sci. USA, 94, 2122–2127. Hazen, B.E. and Bushnell, W.R. (1983) Inhibition of the hypersensitive reaction in barley to powdery mildew by heat shock and cytochalasin B. Physiol. Plant Pathol. 23, 421–438. Heath, M.C. and Heath, I.B. (1971) Ultrastructure of an immune and a susceptible reaction of cowpea leaves to rust infection. Physiol. Plant Pathol. 1, 277–287. Hepler, P.K. (1982) Endoplasmic reticulum in the formation of the cell plate and plasmodesmata. Protoplasma, 111, 121–133. Holub, E.B. (2001) The arms race is ancient history in Arabidopsis, the wildflower. Nat. Rev. Genet. 2, 516–527. Holub, E.B., Beynon, J.L. and Crute, I.R. (1994) Phenotypic and genotypic characterization of interactions between isolates of Peronospora parasitica and accessions of Arabidopsis thaliana. Mol. Plant-Microbe Interact. 7, 223–239. Holub, E.B., Brose, E., Tör, M., Clay, C., Crute, I.R. and Beynon, J.L. (1995) Phenotypic and genotypic variation in the interaction between Arabidopsis thaliana and Albugo candida. Mol. Plant-Microbe Interact. 8, 916–928. Hush, J.M., Hawes, C.R. and Overall, R.L. (1990) Interphase microtubule re-orientation predicts a new cell polarity in wounded pea roots. J. Cell Sci. 96, 47–61. Jones, D.A., Thomas, C.M., Hammond-Kosack, K.E., Balint-Kurti, P.J. and Jones, J.D.G. (1994) Isolation of the tomato Cf-9 gene for resistance to Cladosporium fulvum by transposon tagging. Science, 266, 789–793. Kamoun, S. (2001) Non-host resistance to Phytophthora: novel prospects for a classical problem. Curr. Opin. Plant Biol. 4, 295–300. Kennard, J.L. and Cleary, A.L. (1997) Pre-mitotic nuclear migration in subsidiary mother cells of Tradescantia occurs in G1 of the cell cycle and requires F-actin. Cell Motil. Cytoskeleton, 36, 55–67. Klahre, U., Friederich, E., Kost, B., Louvard, D. and Chua, N.-H. (2000) Villin-like actin-binding proteins are expressed ubiquitously in Arabidopsis. Plant Physiol. 122, 35–47. Kobayashi, Y., Kobayashi, I., Funaki, Y., Fujimoto, S., Takemoto, T. and Kunoh, H. (1997a) Dynamic re-organization of microfilaments and microtubules is necessary for the expression of nonhost resistance in barley coleoptile cells. Plant J. 11, 525–537. Kobayashi, I., Kobayashi, Y. and Hardham, A.R. (1994) Dynamic reorganization of microtubules and microfilaments in flax cells ß Blackwell Publishing Ltd, The Plant Journal, (2003), 33, 775–792 791 during the resistance response to flax rust infection. Planta, 195, 237–247. Kobayashi, I., Kobayashi, Y. and Hardham, A.R. (1997b) Inhibition of rust-induced hypersensitive response in flax cells by the microtubule inhibitor oryzalin. Aust. J. Plant Physiol. 24, 733–740. Kobayashi, I., Kobayashi, Y., Yamaoka, N. and Kunoh, H. (1992) Recognition of a pathogen and a non-pathogen by barley coleoptile cells. Part III. Responses of microtubules and actin filaments in barley coleoptile cells to penetration attempts. Can. J. Bot. 70, 1815–1823. Koch, E. and Slusarenko, A. (1990) Arabidopsis is susceptible to infection by a downy mildew fungus. Plant Cell, 2, 437–445. Köhler, R.H., Zipfel, W.R., Webb, W.W. and Hanson, M.R. (1997) The green fluorescent protein as a marker to visualize plant mitochondria in vivo. Plant J. 11, 613–621. Kost, B., Spielhofer, P. and Chua, N.-H. (1998) A GFP-mouse talin fusion protein labels plant actin filaments in vivo and visualizes the actin cytoskeleton in growing pollen tubes. Plant J. 16, 393–401. Kunoh, H., Aist, J.R. and Hayashimoto, A. (1985) The occurrence of cytoplasmic aggregations induced by Erysiphe pisi in barley coleoptile cells before the host cell walls are penetrated. Physiol. Plant Pathol. 26, 199–207. Lazo, G.R., Stein, P.A. and Ludwig, R.A. (1991) A DNA transformation-competent Arabidopsis genomic library in Agrobacterium. Biotechnology, 9, 963–967. Leckie, C.P., Callow, J.A. and Green, J.R. (1995) Re-organization of the endoplasmic reticulum in pea leaf epidermal cells infected by the powdery mildew fungus Erysiphe pisi. New Phytol. 131, 211–221. Mano, S., Hayashi, M. and Nishimura, M. (1999) Light regulates alternative splicing of hydroxypyruvate reductase in pumpkin. Plant J. 17, 309–320. Marc, J., Granger, C.L., Brincat, J., Fisher, D.D., Kao, T., McCubbin, A.G. and Cyr, R.J. (1998) A GFP-MAP4 reporter gene for visualizing cortical microtubule rearrangements in living epidermal cells. Plant Cell, 10, 1927–1939. Mauch-Mani, B. and Slusarenko, A.J. (1994) Systemic acquired resistance in Arabidopsis thaliana induced by a predisposing infection with a pathogen isolate of Fusarium oxysporum. Mol. Plant-Microbe Interact. 7, 378–383. McCann, R.O. and Craig, S.W. (1997) The I/LWEQ module: a conserved sequence that signifies F-actin binding in functionally diverse proteins from yeast to mammals. Proc. Natl. Acad. Sci. USA, 94, 5679–5684. McLusky, S.R., Bennett, M.H., Beale, M.H., Lewis, M.J., Gaskin, P. and Mansfield, J.W. (1999) Cell wall alterations and localized accumulation of feruloyl-30 -methoxytyramine in onion epidermis at sites of attempted penetration by Botrytis allii are associated with actin polarisation, peroxidase activity and suppression of flavonoid biosynthesis. Plant J. 17, 523–534. Mitsuhashi, N., Shimada, T., Mano, S., Nishimura, M. and HaraNishimura, I. (2000) Characterization of organelles in the vacuolar-sorting pathway by visualization with GFP in tobacco BY-2 cells. Plant Cell Physiol. 41, 993–1001. Murashige, T. and Skoog, F. (1962) A revised medium for rapid growth and bioassays with tobacco tissue cultures. Physiol. Plant. 15, 473–497. Muskett, P.R., Kahn, K., Austin, M.J., Moisan, L.J., Sadanandom, A., Shirasu, K., Jones, J.D.G. and Parker, J.E. (2002) Arabidopsis RAR1 exerts rate-limiting control of R-gene-mediated defenses against multiple pathogens. Plant Cell, 14, 979–992. 792 Daigo Takemoto et al. Nebenführ, A., Frohlick, J.A. and Staehelin, L.A. (2000) Re-distribution of Golgi stacks and other organelles during mitosis and cytokinesis in plant cells. Plant Physiol. 124, 135–151. Nebenführ, A., Gallagher, L.A., Dunahay, T.G., Frohlick, J.A., Mazurkiewicz, A.M., Meehl, J.B. and Staehelin, L.A. (1999) Stop-and-go movements of plant Golgi stacks are mediated by the acto-myosin system. Plant Physiol. 121, 1127–1141. Nick, P. (1999) Signals, motors, morphogenesis – the cytoskeleton in plant development. Plant Biol. 1, 169–179. Niwa, Y., Hirano, T., Yoshimoto, K., Shimizu, M. and Kobayashi, H. (1999) Non-invasive quantitative detection and applications of non-toxic, S65T-type green fluorescent protein in living plants. Plant J. 18, 455–463. Parker, J.E., Feys, B.J., van der Biezen, E.A., Noël, L., Aarts, N., Austin, M.J., Botella, M.A., Frost, L.N., Daniels, M.J. and Jones, J.D.G. (2000) Unravelling R-gene-mediated disease resistance pathways in Arabidopsis. Mol. Plant Pathol. 1, 17–24. Parker, J.E., Szabo, V., Staskawicz, B.J., Lister, C., Dean, C., Daniels, M.J. and Jones, J.D.G. (1993) Phenotypic characterization and molecular mapping of the Arabidopsis thaliana locus RPP5, determining disease resistance to Peronospora parasitica. Plant J. 4, 821–831. Penninckx, I.A.M.A., Eggermont, K., Terras, F.R.G., Thomma, B.P.H.J., De samblanx, G.W., Buchala, A., Métraux, J.-P., Manners, J.M. and Broekaert, W.F. (1996) Pathogen-induced systemic activation of a plant defensin gene in Arabidopsis follows a salicylic acid-independent pathway. Plant Cell, 8, 2309–2323. Quader, H. and Schnepf, E. (1986) Endoplasmic reticulum and cytoplasmic streaming: fluorescence microscopical observations in adaxial epidermis cells of onion bulb scales. Protoplasma, 131, 250–252. Ridge, R.W., Uozumi, Y., Plazinski, J., Hurley, U.A. and Williamson, R.E. (1999) Developmental transitions and dynamics of the cortical ER of Arabidopsis cells seen with green fluorescent protein. Plant Cell Physiol. 40, 1253–1261. Roetschi, A., Si-Ammour, A., Belbahri, L., Mauch, F. and MauchMani, B. (2001) Characterization of an Arabidopsis–Phytophthora pathosystem: resistance requires a functional PAD2 gene and is independent of salicylic acid, ethylene and jasmonic acid signalling. Plant J. 28, 293–305. Ryals, J.A., Neuenschwander, U.H., Willits, M.G., Molina, A., Steiner, H.-Y. and Hunt, M.D. (1996) Systemic acquired resistance. Plant Cell, 8, 1809–1819. Schena, M., Lloyd, A.M. and Davis, R.W. (1991) A steroid-inducible gene expression system for plant cells. Proc. Natl. Acad. Sci. USA, 88, 10421–10425. Škalamera, D. and Heath, M.C. (1996) Cellular mechanisms of callose deposition in response to fungal infection or chemical damage. Can. J. Bot. 74, 1236–1242. Škalamera, D. and Heath, M.C. (1998) Changes in the cytoskeleton accompanying infection-induced nuclear movements and hypersensitive response in plant cells invaded by rust fungi. Plant J. 16, 191–200. Staiger, C.J. (2000) Signaling to the actin cytoskeleton in plants. Annu. Rev. Plant Physiol. Plant Mol. Biol. 51, 257–288. Stössel, P., Lazarovits, G. and Ward, E.W.B. (1980) Penetration and growth of compatible and incompatible races of Phytophthora megasperma var. sojae in soybean hypocotyl tissues differing in age. Can. J. Bot. 58, 2594–2601. Takemoto, D., Maeda, H., Yoshioka, H., Doke, N. and Kawakita, K. (1999) Effect of cytochalasin D on defense responses of potato tuber discs treated with hyphal wall components of Phytophthora infestans. Plant Sci. 141, 219–226. Thomma, B.P.H.J., Eggermont, K., Penninckx, I.A.M.A., MauchMani, B., Vogelsang, R., Cammue, B.P.A. and Broekaert, W.F. (1998) Separate jasmonate-dependent and salicylate-dependent defense-response pathways in Arabidopsis are essential for resistance to distinct microbial pathogens. Proc. Natl. Acad. Sci. USA, 95, 15107–15111. Tomiyama, K. (1956) Cell physiological studies on the resistance of potato plant to Phytophthora infestans. Ann. Phytopathol. Soc. Jpn. 21, 54–62. Tomiyama, K., Sato, K. and Doke, N. (1982) Effect of cytochalasin B and colchicine on hypersensitive death of potato cells infected by incompatible race of Phytophthora infestans. Ann. Phytopathol. Soc. Jpn. 48, 228–230. Ueda, K., Matsuyama, T. and Hashimoto, T. (1999) Visualization of microtubules in living cells of transgenic Arabidopsis thaliana. Protoplasma, 206, 201–206. Vleeshouwers, V.G.A.A., van Dooijeweert, W., Govers, F., Kamoun, S. and Colon, L.T. (2000) The hypersensitive response is associated with host and non-host resistance to Phytophthora infestans. Planta, 210, 853–864. Xu, J.-R., Staiger, C.J. and Hamer, J.E. (1998) Inactivation of the mitogen-activated protein kinase Mps1 from the rice blast fungus prevents penetration of host cells but allows activation of plant defense responses. Proc. Natl. Acad. Sci. USA, 95, 12713–12718. Zhou, N., Tootle, T.L. and Glazebrook, J. (1999) Arabidopsis PAD3, a gene required for camalexin biosynthesis, encodes a putative cytochrome P450 monooxygenase. Plant Cell, 11, 2419–2428. ß Blackwell Publishing Ltd, The Plant Journal, (2003), 33, 775–792