Physiological and Molecular Plant Pathology (2001) 58, 209±228 doi:10.1006/pmpp.2001.0330, available online at http://www.idealibrary.com on

Inhibition of Blumeria graminis germination and germling development within colonies of oat mildew

T. L.W. CARVER1*, P. C. ROBERTS1,B.J.THOMAS1 and M. F. LYNGKáR2 1Institute of Grassland and Environmental Research, Aberystwyth, Ceredigion SY23 3EB, U.K. and 2Risù National Laboratory, DK-4000 Roskilde, Denmark

(Accepted for publication April 2001)

Germination by Blumeria graminis DC Speer €. spp. avenae, hordei and tritici, was greatly suppressed when conidia fell within colonies of €. spp. avenae or hordei established on susceptible oat or barley, respectively. On healthy oat or barley, and when distant from powdery mildew colonies, all €. spp. formed normal appressoria. This was also true when conidia germinated within established barley mildew colonies. Within barley mildew colonies, appressoria of f. sp. hordei penetrated epidermal cells (formed haustoria) more frequently than appressoria distant from colonies. Similarly, €. spp. avenae and tritici, normally unable to infect barley, frequently penetrated epidermal cells subtending established barley mildew colonies. Thus, colony establishment induced barley epidermal cell accessibility, even to non-pathogenic €. spp. In contrast, when all three €. spp. germinated within established oat mildew colonies, most formed abnormal, hypha-like germ tubes. Since they did not form appressoria, and were thus unable to attempt penetration, it was impossible to determine whether oat mildew colonies induced accessibility of underlying oat epidermal cells. However, when super®cial structures of established colonies were removed, germlings of all €. spp. formed appressoria freely on cells containing oat mildew colony haustoria. Furthermore, these cells showed high level induced accessibility not only to f. sp. avenae but also to the normally non-pathogenic €. spp. This indicated that factors disrupting germination and further development by conidia lying within oat mildew colonies were produced from the super®cial colony structures and not by haustorium- containing plant cells. The factors appear to have limited mobility. *c 2001 Academic Press

Keywords: Blumeria graminis; Erysiphe graminis; oat mildew; barley mildew; germination; germ tubes; biocontrol.

INTRODUCTION the appressorium. As it grows epiphytically, the elongat- ing appressorial germ tube forms a single septum and itself Blumeria graminis DC Speer (syn. Erysiphe graminis DC) recognizes host surface features [4±6, 17]. This recognition causes powdery mildew of cereals and other grasses. The leads to swelling of the distal cell and eventually to disease is spread mainly by wind-blown asexual conidia. di€erentiation of a hooked appressorial lobe at the tube Following deposition on a host, conidia germinate and apex (Fig. 1). At 208C, appressoria are formed by 10± germling morphogenesis proceeds through a series of 12 h after inoculation. Occasionally, ®rst- or later-formed distinct steps. Emergence of the primary germ tube germ tubes fail to contact, or recognize contact with the (PGT) provides the ®rst visible sign of germination (within 30±120 min). The PGT remains short (approx. leaf, in which case they remain as short, non-functional 5±10 mm) and aseptate, and is thought to have at least ``subsidiary germ tubes'' [7, 26, 38]. In this case, reserves three functions, namely, attachment to the host surface permitting, a subsequent germ tube will take on the PGT [5±7, 11, 26], gaining access to host water [2], and functions and the next-formed germ tube will elongate. recognising host surface features [4, 6, 7, 38]. The features Similarly, elongated germ tubes sometimes fail to contact, recognized probably include hydrophobicity and cutin or respond to, the host surface, in which case they remain and cellulose breakdown products released by fungal as undi€erentiated, hypha-like structures incapable of enzymes [5, 6, 10, 14, 32, 35, 36]. This recognition engages infection [4, 5, 38]. processes that lead to elongation of a second germ tube When a functional appressorium is formed, an infec- [4, 5, 38] that emerges shortly after the PGT [5, 6, 26, 38]. tion peg emerges from beneath the appressorial lobe and Elongation of the second germ tube is pre-requisite to it attempts to penetrate the underlying epidermal cell (12± eventually di€erentiating a specialized infection structure, 16 h after inoculation) [1]. The attacked cell responds and papilla deposition is initiated. However, in a com- * To whom all correspondence should be addressed. patible relationship, the infection peg often penetrates the 0885-5765/01/050209+20 $35.00/00 *c 2001 Academic Press 210 T. L. W. Carver et al.

(a) haustoria (with increasing numbers in each generation) are formed nightly. By 4±5 days the ®rst conidiophores PGT develop and sporulation commences shortly afterwards. Even in compatible host±B. graminis interactions where S App some primary penetration attempts succeed, many fail to C penetrate the papilla defence. It is now well established that the ability of B. graminis to penetrate cereal epidermal cells is dramatically a€ected by the outcome of prior attacks by the fungus [8, 20, 21, 23, 24, 27±29, 33, 34]. This has implications for the ®eld situation where individual H cells may be confronted by sequential attacks as conidia are deposited continually from the aerial pool. Thus, if an 10 mm initial attack is successful, the cell shows induced accessibility (sensu Ouchi et al.[34]) so that later attacks on that cell will almost invariable succeed. By contrast, (b) failed initial attacks induce ``inaccessibility'' (sensu Kunoh et al.[21]) within the cell so that subsequent attacks almost PGT C invariably fail. While these induced changes are essen- tially localized phenomena, being expressed most strongly in the single cell directly subjected to earlier attack, there S is some transmission of e€ects to immediately adjacent epidermal cells although the e€ects are often undetectable App at two cells distance [8, 21, 23, 27±29, 33]. Lyngkjñr and Carver's previous studies [8, 27±29] involved double inoculation procedures whereby the ®rst, ``inducer'' inoculum was allowed to develop and then its 10 mm super®cial fungal structures (conidia, germ tubes, H hyphae) were removed before the second, ``challenge'' inoculum was applied. The inducer inoculum was given a FIG. 1. (a and b) Light and LTSEM micrographs, respectively, maximum of 48 h incubation before removal of super®cial of ``normal'' B. graminis f. sp. avenae germlings 24 h after inoculation. Each conidium (c) has formed a short, aseptate structures i.e. where primary penetration had succeeded, primary germ tube (PGT) and a second appressorial germ tube a primary haustorium was formed but no hyphal (App) that has elongated (to approx. 40 mm), swollen and has a appressoria or second generation haustoria were initiated. distinct, hooked apical lobe (H). A single septum (S) is evident The success (or failure) of challenge attack was then within the appressorial germ tube (approx. 10±15 mmfromthe determined in relation to the presence of inducer primary conidium). haustoria or papillae remaining as intracellular evidence of successful or failed inducer attack, respectively. response site, enters the cell and an immature primary We undertook the present work to extend Lyngkjñr and haustorium is formed (15±24 h). The haustorium then Carver's earlier studies [8, 27±29]. Here, however, we develops digitate processes from each end of the aimed to allow colonies from inducer inoculum to develop ellipsoidal haustorial body and absorbs nutrients from second generation haustoria before applying challenge the infected cell [1, 3, 18]. These are exported to the inoculum. We intended that germling appressoria of the appressorium, hyphae emerge from the appressorial germ challenge inoculum would attack host cells at the time tube or from the mother conidium, and an epiphytic that inducer colonies were initiating their third generation hyphal weft grows at accelerating rate over the leaf haustoria. Thus, our intention was to determine whether surface [3, 18]. After 2±3 days, nutrients supplied by the the presence of multiple haustoria (second and third growing primary haustorium become insucient to generation) produced by a developing colony enhanced support the young colony, and a second generation of the expression of induced accessibility in nearby, unat- hyphal appressoria is initiated in response to darkness tacked cells. during the daily cycle [3, 18]. These hyphal appressoria We executed our ®rst experiments using oat (Avena are usually sited either on the cell containing the primary sativa L.) inoculated with a compatible isolate of haustorium or within one or two cells distance of it. B. graminis f. sp. avenae Marchal. However, we found it Penetrations from these appressoria form secondary impossible to gather sucient data to ful®l our objective. haustoria (usually between two and six) and hyphal Where challenge conidia were deposited within the growth accelerates further. Successive generations of boundaries of an established f.sp. avenae colony derived Inhibition of B. graminis development 211 from inducer inoculation, very few of these conidia To produce disease-free experimental plants, seeds were seemed to germinate. Furthermore, very few of those sown singly in paper tubes (15  120 mm) containing that did germinate formed an appressorium. The soil : peat : sand (60 : 30 : 10; v : v). Plants were grown to majority of germlings produced highly abnormal full expansion of the second-formed leaf (approx. 18 days) hypha-like germ tubes and did not appear to attempt in a spore-proof glasshouse under natural lighting penetration of the host cells. conditions and a minimum temperature of 128C. The work reported here was therefore redirected towards exploring the e€ects that established colonies Experimental designs and procedures might have on germination and germling development by B. graminis conidia deposited within established Double inoculation experiments: established colonies in place. B. graminis colonies. We investigated these e€ects with All of these experiments involved using ®rst a low density respect to f. sp. avenae colonies on susceptible oats and f. inoculation to establish young colonies on test leaves and sp. hordei colonies on susceptible barley. then, 72 h later, applying a second, high density inoculation. In this way, conidia of the second inocu- lation were deposited both within established colonies where they lay touching or between the super®cial hyphae, or distant from colonies when they lay on cells MATERIALS AND METHODS that had no contact with hyphae of established colonies. For the ®rst inoculation, in all cases nine leaves of Selma Pathogens and plants or Maldwyn oat, or of Pallas barley were inoculated with An isolate of B. graminis f. sp. avenae race 5, was multiplied conidia of the appropriate f. sp. (avenae or hordei, on plants of oat cv. Selma grown in a spore-proof respectively). Paper tubes containing the plants were glasshouse under natural lighting conditions and mini- laid on a bench, and their second-formed leaves were mum temperature of 128C. Similarly, isolate CC1 of arranged adaxial surface up (and held ¯at using small B. graminis f. sp. hordei and an isolate of B. graminis f. sp. weights at leaf apex and base), before placing a spore tritici (originally supplied by A. Daniels, Aventis, U.K.), settling tower over them. Conidia were blown from were multiplied on barley cv. Pallas and wheat cv. infected leaves into the tower and a slide placed among Riband, respectively. One day before conidia were the leaves was used to trap conidia and monitor spore required for experimentation, infected leaves were shaken density which was adjusted to approx. 2±5 conidia mm2. to remove older conidia and ensure a supply of vigorous The inoculated plants and four additional healthy plants young conidia for inoculum. were then transferred to a growth cabinet at 208Cand For all experiments, fully expanded second-formed under a 12 h light/12 h dark cycle, with 200 mmol leaves of cereal plants were used. Studies of oat (A. sativa m2 s1 photon ¯ux density during the light period. L.) used cvs Selma and Maldwyn. Neither has known The ®rst light period commenced immediately after major gene resistance to B. graminis f. sp. avenae race 5. inoculation. Studies of barley (Hordeum vulgare L.) used cv. Pallas, Under these conditions, previous studies [3, 9, 12] which is susceptible to B. graminis f. sp. hordei isolate CC1. allowed us to anticipate the pattern of colony develop- Since the second leaves of these cvs are susceptible to ment. Following establishment of a functional primary virulent fungal isolates, penetration attempts from many haustorium at around 20 h after inoculation, secondary appressoria are successful and relatively few attacks fail in hyphal elongation and branching was expected to association with papilla deposition by living epidermal accelerate up to the dark period beginning 60 h after cells, and few result in death of attacked cells ([8], inoculation, when, in the 60±72 h dark period, hyphal Lyngkjñr and Carver, unpublished). appressoria and then second generation haustoria would Neither oat cv. nor Pallas barley is susceptible to be initiated. One of the nine inoculated leaves was cut, B. graminis f. sp. tritici, Pallas is not susceptible to B. graminis ®xed, cleared and microscopically examined 71 h after f. sp. avenae, and the oat cvs are not susceptible to inoculation to con®rm that colonies were developing as B. graminis f. sp. hordei. As expected from previous studies expected. When this was con®rmed, four of the inocu- [13], preliminary observations suggested that failure of lated leaves were ®xed 72 h after inoculation for infection by inappropriate €. spp. was explained by a reference. combination of two factors: either papilla responses At 72 h, a second inoculation was applied to the prevented penetration or attacked cells died rapidly in remaining four inoculated leaves, and the second-formed response to attack. Importantly, even on inappropriate leaves of the four healthy plants were inoculated simulta- cereal spp., conidia of all fungal €. spp. appeared to neously to provide a control. The second inoculation was germinate and develop normally up to the stage of adjusted to approx. 50 conidia mm2. According to appressorial di€erentiation. requirement, conidia for the second inoculation were 212 T. L. W. Carver et al. either of the f. sp. appropriate to the host (i.e. the same second inoculum development within established colo- isolate as used for establishing colonies via the ®rst nies. We needed, therefore, to stain specimens. inoculation) or of an inappropriate f. sp. normally unable The ®rst experiment involved double inoculating to infect the test leaves. Following the second inoculation, Selma oat with B. graminis f. sp. avenae, leaving leaves were returned to the growth cabinet and incubated established colonies from the ®rst inoculation in place. for a further 24 h before all were ®xed. Here, staining was achieved by placing a smear of Trypan blue in lactoglycerol (0.1 g per 100 ml) on a Double inoculation experiments: established colonies removed. coverslip, lowering the adaxial (inoculated) surface of the Thirteen second-formed leaves of Selma oat were leaf segment onto the coverslip, and then inverting the 2 inoculated with approx. 2 conidida mm of B. graminis mount onto a microscope slide smeared with lactogly- f. sp. avenae race 5. These, and 12 healthy (uninoculated) cerol. This procedure was devised to allow staining leaves, were incubated as described above, but for 96 h, without displacing loosely attached germlings. We used although one inoculated leaf was ®xed, cleared and this to gather data on the developmental characteristics microscopically examined 95 h after inoculation. Micro- of germlings formed by conidia of the second inoculum scopy of this leaf con®rmed the anticipation [3] that the in relation to their proximity to established colonies. third generation of haustoria had been initiated by this However, we found that even with care some ungermi- time. At 96 h after the ®rst inoculation, the adaxial nated conidia were displaced; they could be seen moving surface of all leaves (including the healthy ones) was over relatively long distances if specimens were viewed painted with cellulose acetate/acetone solution which was immediately after mounting when the liquid media were allowed to dry (approx. 2±5 min) before the dried acetate ¯owing over segments. Therefore, for this experiment we ®lm was peeled away. This removed all super®cial fungal had limited con®dence in data for germination frequency structures due to the ®rst inoculation (colony hyphae, by conidia of the second inoculum in relation to their failed germlings and conidia) and the epicuticular leaf proximity to established colonies. waxes from inoculated and healthy leaves [10, 27]. As To allow assessment of germination frequency as well expected [27], epidermal cells containing haustoria as the morphological characteristics of germlings formed formed by colonies on leaves treated with the ®rst by the second inoculum, we devised a staining technique inoculum remained alive (as judged by lack of cyto- that caused no detectable displacement of ungerminated plasmic granulation and whole-cell auto¯uorescence). conidia. This was used for all further experiments. As Four leaves that had received the ®rst inoculum were then before, leaves were ®xed, bleached, and cleared, but inoculated for a second time using 50 conidia mm2 of either f. sp. avenae, f. sp. hordei or f. sp. tritici, and di€erent dishes containing the segments supported on tissue paper sets of four healthy leaves were also inoculated simul- were held on a slant to avoid accumulation of free taneously with the di€erent €. spp. All were incubated for liquids around segments and the possibility that free a further 36 h before ®xation. liquid would move onto the segment surface. To stain, segments supported on slanted dishes were sprayed very lightly using a hand-pumped atomizer (Schott, Germany) from a distance of approx. 40 cm with Preparation for light microscopy alcoholic methyl blue stain (0.2 g methyl blue in As described previously (e.g. [13]) leaves were ®xed and 100 ml 95 % ethanol). A small amount of spray (six cleared by a procedure that avoids displacing ungermi- pumps) was applied and then spraying was suspended nated conidia and loosely attached germlings. Brie¯y, for 1 min to allow the ethanol to evaporate. In this way, central 30 mm segments of leaves were excised, and laid no visible droplets of liquid accumulated. A further inoculated surface up on tissue paper pads moistened spray was then applied. This process was repeated until with 3 : 1 ethanol : glacial acetic acid (v : v) to ®x and microscope inspection showed colonies, germlings and bleach the leaves. They were then transferred to pads ungerminated conidia to be stained suciently (usually moistened with water and subsequently to pads mois- three cycles). Repeated observation (before and after tened with lactoglycerol (lactic acid : glycerol : water, spray staining, without coverslip) of spores located 1 : 1 : 1, v : v) to clear. For studies of fungal development distantly from established colonies, showed that the where leaves are ®xed before large colonies are allowed to procedure did not displace ungerminated conidia. develop, it is possible to attain good resolution of Following staining, specimens were mounted without a pathogen and host using unstained leaves mounted coverslip and viewed by transmitted light and di€eren- without coverslips, and viewed with no-coverslip objec- tial interference contrast microscopy using 20  or tives. However, because of di€raction around hyphae of 40  objectives. Observations were made immediately colonies established by the ®rst inoculum, it was because although staining remained good for approx. impossible to use this approach to resolve details of 24 h, it faded thereafter. Inhibition of B. graminis development 213 Data collection apical lobe. Note was also made of whether appressoria were associated with successful penetration as evidenced Double inoculation experiments: established colonies in place. by the presence of a haustorium. Abnormal germlings Our aim in these experiments was to determine the e€ect were those which had either failed to form an elongated colonies established from the ®rst inoculation might have germ tube (at least 15 mm), or where the elongated germ on conidial germination and the development of germl- tube had not di€erentiated into an appressorial germ ings from the second inoculation. tube. On double-inoculated leaf segments ®xed 24 h after For the ®rst experiment where coverslips were ®tted, on application of the second inoculum, observations were each control leaf (inoculated once, 24 h before ®xation) made for 50 germinated conidia that lay on leaf 50 randomly selected, germinated conidia were epidermal cells located ``distant'' from the nearest colony, examined. Conidia were classi®ed as germinated if at and germlings were classi®ed in the same way as on least one short germ tube had been produced. For each of controls. Distant cells were de®ned as being at least four these germlings, note was taken of whether they had epidermal cells away from the nearest cell in contact with developed abnormally or had developed normally to hyphae belonging to colonies formed by the ®rst di€erentiate an appressorial germ tube (illustrated in inoculum. Similar classi®cations were made for 50 Fig. 1). Appressorial germ tubes were recognized by: the germlings arising from conidia located within the outer presence of a single septum within the tube, swelling ``boundaries'' of colonies established by the ®rst inocu- towards the tube apex, and the presence of a hooked lum. The colony boundary was de®ned by an imaginary line joining the tips of a colony's leader hyphae (Fig. 2). Of course, on double inoculated leaves, some germlings (a) Colony 72 h after inoculation deriving from the ®rst inoculation failed to penetrate and Max. longitudinal distance C establish colonies, and we could not distinguish between

Max. lateral distance these and germlings from the second inoculation. Some of 100 µm these failed germlings would inevitably be included in our samples. However, since the density of the second (b) Colony 96 h after inoculation inoculation (appox. 50 conidia mm2) was 10±25 times greater than the ®rst (approx. 2±5 conidia mm2), and La1 since many germlings from the ®rst inoculation formed colonies, the great majority of data would necessarily relate to germlings derived from the second inoculation. Lo1 Lo2 Therefore, we introduced only small errors due to La2 100 µm inclusion of observations of germlings from the ®rst inoculation. (c) Inner and outer zones of 96 h colony We wished to gauge the growth made by colonies in the period 72±96 h after inoculation i.e. from the time of applying second inoculum to the time of ®xation for double inoculated leaves. Drawings of two ``typical'' B. graminis f. sp. avenae colonies are shown in Fig. 2(a) and (b) (for 72 h and 96 h colonies, respectively). As measures of colony size, the maximum longitudinal and Outer zone lateral distances from tip to tip of the longest hyphae Inner zone [Fig. 2(a) and (b)] were measured for 10 colonies on each FIG. 2. Drawings of B. graminis f. sp. avenae colonies ®xed 72 h leaf segment ®xed 72 h after inoculation, and for 10 (a) and 96 h (b) and (c) after inoculating oat cv. Selma. (a) colonies ®xed 96 h after inoculation. For the 96 h Positioning of co-ordinates for measurement of maximum segments, measurements were also made of the maximum longitudinal and lateral measurements. C ˆ conidium. (b) Positioning of co-ordinates marking the last branches on leader longitudinal and lateral distances between the last hyphae. Measurement of Lo1 to Lo2 gives the longitudinal branches on leader hyphae [Fig. 2(b)]. From these distance between last branches; La1 to La2 gives lateral distance measurements, it was deduced that the lengths and between last branches. (c) The inner and outer zones of a 96 h breadths of 72 h colonies were approximately equivalent colony [same as shown in (b)]. The inner zone lies within an to the last-branch length and breadth distances seen in imaginary boundary line linking the points at which the last 96 h colonies. Thus, when observing 96 h colonies, the hyphal branch occurred on the colony's major runner hyphae. The outer zone (shaded) lay between the inner boundary and position of the last branch on leader hyphae o€ered an the outer boundary which was de®ned by an imaginary line approximate, but reasonable, indication of the location of joining the tips of the colony's major hyphal branches. leader hyphal tips at 72 h when the second inoculum was 214 T. L. W. Carver et al. applied. This information allowed us to modify data on epidermal cells at least four cells away from the nearest collection criteria for subsequent experiments. cell containing a haustorium formed by colonies arising In remaining experiments where the fungus was spray- from the ®rst inoculum (distant cells), and 25 germlings stained, additional data were collected. As described on distant cells were scored for their developmental above, on controls (inoculated once, 24 h before ®xation) characteristics. Similarly where conidia of the second 50 germlings were classi®ed for normality of develop- inoculation lay on epidermal cells containing haustoria ment, but in addition, 100 randomly selected conidia formed by colonies of the ®rst inoculation, germination were examined to determine whether they had germi- was determined for 50 conidia and germling development nated. Similarly, on double-inoculated leaves ®xed 24 h was assessed for 25 germinated conidia. after the second inoculation, 50 germlings lying on As described above, mean data from each leaf segment epidermal cells distant from colonies were classi®ed for were calculated as percentages that were transformed normality of development, and 100 conidia on such cells prior to analysis of variance. were examined for germination. In relation to established colonies, data were collected from two di€erent zones within colonies. The inner zone was de®ned by an Low temperature scanning electron microscopy (LTSEM) imaginary boundary line linking the points at which the To complement light microscopy and aid image last hyphal branch occurred on the colony's major runner interpretation, sub-samples of double inoculated leaves hyphae [Fig. 2(c)]. The outer zone lay between the inner were examined by LTSEM. Thus, when the central boundary and the outer boundary which was, as before, 30 mm leaf segments were excised and ®xed for light de®ned by an imaginary line joining the tips of a colony's microscopy, the adjacent basipetal 10 mm leaf blade major hyphal branches [Fig. 2(c)]. On each leaf segment, segment from some samples was taken for LTSEM. many randomly selected colonies were examined and Following established procedure [11], these segments data were collected from within each zone until we had were attached to a copper SEM stub with colloidal accumulated observations from 100 conidia to determine graphite and frozen rapidly by introducing the stub to a whether they had germinated, and from 50 germinated pre-cooled stub holder (approx. 1908C) under an conidia for normality of germling development. argon ¯ush, and transferring the assembly to the pre- As considered above, on double inoculated leaves, only cooled (1908C) stage of an Emscope SP2000A sputter- small errors would have been introduced due to inclusion cryo system. Cryo-®xation in this way avoided ice crystal of data deriving from the ®rst inoculum in the data sets. formation on specimens. The specimen was then sputter For all leaf segments ®xed 24 h after the second coated with gold before detailed observation on the cold inoculation and in all experiments, mean percentages stage (approx. 1608C) of a JEOL JSM 840A SEM, were calculated, as appropriate, for spore germination, with 3.0or5.0 kV electron accelerating voltage. SEM abnormal germling development, and haustorium for- images were recorded using a digitizer board and mation. Data sets were gathered from each location on software (SemAfore2, JEOL) running under a Win- double inoculated leaves (distant from colonies, and, as dows2 operating system. The ®gures presented here are appropriate, within colony outer and inner zones) and for from digitized images enhanced using Adobe Photo- controls inoculated once 24 h before ®xation. Percentages shop2 to maximize details of areas of interest. were transformed to arcsine square roots (transformed value ˆ 180/p  arcsine [Z/(100)]) to normalize data and stabilize the variance throughout the data range, before being subjected to analysis of variance (using RESULTS Genstat [15]). Selma oat double-inoculated with f. sp. avenae, coverslips applied Double inoculation experiments: established colonies removed. Our aim here was to determine whether any e€ects of the Data for spore germination will not be described because ®rst inoculation on subsequent performance of conidia applying coverslips to specimens was seen to displace and germlings arising from the second inoculation, many ungerminated spores. Fig. 3 shows that deposition depended: (1) upon the presence of colony hyphae within the boundary of an established colony had a derived from the ®rst inoculation; or (2) whether the dramatic e€ect on germling development. On control presence of haustoria within epidermal cells was sucient leaves (inoculated only once, 24 h before ®xation) and to mediate these e€ects. when germlings were distant from established colonies on On each control leaf segment, data to assess germina- double-inoculated leaves, the great majority of germlings tion and normality of germling development were developed normal appressoria. By contrast, when conidia gathered as described above. On leaves inoculated were located within the boundary of an established twice, germination was determined for 50 conidia located colony, more than 70 % of germlings were abnormal Inhibition of B. graminis development 215

100 apex. When germlings were within the boundary of (a) established colonies, they frequently made direct contact 80 b with hyphae of the established colony [Fig. 4(a), (b), (d) and (e)]. Contact could be via the conidium, the germ 60 tubes, or both. However, abnormal germlings were

(%) formed even where no direct contact was made 40 [Fig. 4(c) and (f)]. On controls and when germlings were distant from 20 a a established colonies on double inoculated leaves, approx. 15±30 % of germlings successfully penetrated leaf epi- 0 dermal cells to form a haustorium [Fig. 3(b)]. By 50 contrast, when germlings were located within the (b) boundary of an established colony, less than 5 % of 40 germlings formed haustoria. a 30 Experiments where specimens were viewed without coverslips, (%) 20 ab and hyphae of established colonies remained in place Oats inoculated ®rst with f. sp. avenae and secondly with €. 10 b spp. avenae, hordei or tritici

0 Germination. When spray stained specimens were mounted without coverslips to avoid displacing ungermi-

Within nated conidia we felt con®dent that data gathered to Control Distant describe spore germination were valid. Here, therefore, Conidium location relative to colony we considered percentage germination of conidia located on control leaves, and, where leaves bore colonies FIG. 3. Development of B. graminis f. sp. avenae germlings on oat established from the ®rst inoculation, we considered the cv. Selma second-formed leaves 24 h after inoculation onto germination of conidia deposited distant from colonies healthy control leaves or onto leaves bearing B. graminis f. sp. avenae colonies derived from a ®rst inoculation 72 h earlier. All and lying within the outer and inner zones of colonies material was ®xed 24 h after the second inoculation. (a) [Fig. 2(c)]. Percentage of morphologically abnormal germlings that failed Fig. 5(a) and (b) show that when f. sp. avenae conidia to develop an appressorium. (b) Percentage of germlings with were inoculated onto Maldwyn or Selma oat leaves apparently normal appressoria that penetrated plant epidermal bearing established f. sp. avenae colonies, germination cells successfully to form a haustorium. On controls ( ), data rates were signi®cantly depressed if conidia lay within the are based on counts of 50 randomly selected germlings, while on inner zone of an established colony. Here, germination leaves bearing colonies, counts were made for 50 germlings located distant from colonies ( ) and a further 50 germlings percentages were approximately halved compared with were located within outer colony boundaries (h), on each of locations distant from colonies or on controls inoculated four replicate leaves in all cases. Percentage data were only once. This was true for experiments involving both transformed to angles for analysis of variance. Where columns oat cvs. There was no di€erence between controls and are headed by di€erent letters, LSD tests applied to angle- locations distant from colonies on double-inoculated . transformed data indicated a signi®cant di€erence (P 5 0 05) leaves. In both experiments, there was a tendency for between means within a data set. conidia within the outer zone of colonies to show slightly lower germination than at distant locations or on [Fig. 3(a)]. Sometimes more than one short germ tube controls, but this e€ect was small and signi®cant only was formed, and none elongated, but this was relatively for the experiment with cv. Maldwyn. rare. Most often, conidia that germinated formed a short When f. sp. tritici conidia were inoculated onto PGT and a second germ tube that grew to a length far Maldwyn oat leaves bearing f. sp. avenae colonies, e€ects greater than normal for an appressorial germ tube of location on germination by f. sp. tritici conidia were (Fig. 4). These long tubes failed to form a hooked very similar [Fig. 5(c)] to those seen in Fig. 5(a) and (b). appressorial lobe at their apex. Sometimes they appeared Germination frequency by conidia located within the to be slightly swollen at the point where an appressorial inner zone of f. sp. avenae colonies was about half that seen lobe would be expected (30±50 mm from the conidium), for conidia distant from colonies or on controls, but but the germ tube then narrowed and extended as a germination was not depressed if conidia were within the hypha-like tube with parallel walls and a simple, rounded outer zone of colonies. 216 T. L. W. Carver et al.

(a) (d) Cu Ab Ab Hy Hy C PGT SGT

C PGT C Ab PGT 50 µm 30 µm

(b) (e) Cu

Hy Ab Hy Ab PGT

C C

C' 50 µm 30 µm

(c) (f) Ab Hy Hy

Ab

C C PGT 50 µm 30 µm PGT

FIG. 4. Light (a)±(c) and LTSEM (d)±(f) micrographs showing morphologically abnormal germlings of B. graminis f. sp. avenae growing within the outer boundary of B. graminis f. sp. avenae colonies on Selma oat leaves. Colonies were established from a ®rst inoculation applied 72 h earlier than the second inoculation from which the germlings derive. All material was ®xed 24 h after the second inoculation. Often, germlings derived from the second inoculum had contact with hyphae of established colonies (a), (b), (d) and (e) via either the conidium, germ tubes or both, although in some cases no contact was made (c) and (f). Sometimes, conidia were ungerminated (a) and (b) or had formed only one [bottom spore, (b)] or more short germ tubes. Where a long germ tube was produced, a short germ tube with the appearance of a typical primary germ tube (a)±(d) and (f) was almost invariably visible. [In (e) it is probable that a primary germ tube is hidden beneath the conidium]. Occasionally germlings with a long germ tube had more than one short germ tube (d) in which case one of these made contact with the host surface and had the characteristics of a primary germ tube, while the other(s) was typical of a redundant subsidiary germ tube. Frequently, abnormal long germ tubes became distinctly swollen towards a region approx. 30±50 mm from the conidium before the tube narrowed and continued growth as a hypha-like tube with parallel walls and a simple rounded apex [germling on right in (a), central germling in (b) and (e)], although in some cases no sign of swelling was evident (f). Ab ˆ abnormal long germ tube; C ˆ conidium; Cu ˆ ungerminated conidium; C0 ˆ conidium with short germ tube only; Hy ˆ hypha of established colony; PGT ˆ primary germ tube; SGT ˆ subsidiary germ tube.

For f. sp. hordei, the suppression of germination marked for conidia deposited within the outer zone of associated with deposition within f. sp. avenae colonies f. sp. avenae colonies, although even here a signi®cantly appeared even greater than for conidia of other €. spp. lower percentage germinated than if located distant from [Fig. 5(d)]. When f. sp. hordei conidia were deposited colonies. Interestingly, there was also a small but within the inner zone of colonies, germination was signi®cant reduction in germination for f. sp. hordei reduced by nearly 70 % (to approx. 17 %) compared conidia located distant from f. sp. avenae colonies on with locations distant from the colonies (appox. 53 % double inoculated leaves compared with controls. This germination). Again, suppression of germination was less e€ect was not seen in any other combination. Inhibition of B. graminis development 217

100 (a) (b) 80

a 60 a b 40 a c a a 20 b

0

100 (c) (d) 80 a Conidia that germinated (%) 60 a a b a c 40 b 20 d

0 zone zone zone zone Inner Inner Outer Outer Control Control Distant Distant

Conidium location Conidium location relative to colony relative to colony

FIG. 5. Percentages of B. graminis €. spp. avenae, tritici and hordei conidia that had germinated 24 h after inoculation onto second- formed oat leaves of healthy controls, or onto leaves bearing B. graminis f. sp. avenae colonies derived from a ®rst inoculation 72 h earlier. All material was ®xed 24 h after the second inoculation. Data are for f. sp. avenae conidia on oat cvs Maldwyn (a) and Selma (b), and for €. spp. tritici (c) and hordei (d) on cv. Maldwyn. On controls ( ) data are based on counts of 100 randomly selected conidia, while on leaves bearing colonies, counts were made for 100 conidia located distant from colonies ( ), 100 within the outer zone of colonies ( ), and 100 located within the inner zone of colonies ( ), on each of four replicate leaves in all cases. Percentage data were transformed to angles for analysis of variance. Where columns are headed by di€erent letters, LSD tests applied to angle- transformed data indicated a signi®cant di€erence (P 5 0.05) between means within a data set.

Germling development. Fig. 6(a) and (b) show that when f. of f. sp. avenae colonies, the abnormality frequency was sp. avenae conidia germinated on Maldwyn or Selma oat slightly but signi®cantly higher (approx. 16 %). leaves bearing established f. sp. avenae colonies, a very However, when f. sp. tritici conidia were within the high percentage (474 %) developed abnormally if inner colony zone, the majority of germlings developed conidia lay within the inner zone of an established abnormally (77 %). The abnormalities were similar to colony. As described for the ®rst experiment (Fig. 4), those shown in Fig. 4. abnormal germlings were those that failed to form a long As with e€ects on germination, the e€ects of germling germ tube, or, where a long germ tube was formed, it did location on the frequency of abnormal germling devel- not di€erentiate an appressorium. In general, abnormal- opment appeared greater for f. sp. hordei than other €. ities were infrequent (appox. 4±12 %) for germlings on spp. As in other cases, when f. sp. hordei germlings were on controls or located distant from colonies. They were also control leaves or distant from f. sp. avenae colonies, a very relatively infrequent when germinated conidia lay within low frequency of abnormality was seen [59%, the outer zone of established colonies (518 %). Fig. 6(d)], and when located within the inner zone of When f. sp. tritici germlings developed on Maldwyn oat colonies abnormalities were extremely frequent [82 %, leaves bearing f. sp. avenae colonies, e€ects of location on Fig. 6(d)]. However, in contrast to other combinations, a germling development by f. sp. tritici [Fig. 6(c)] were very relatively high abnormality frequency [appox. 49 %, similar to those seen in Fig. 6(a) and (b). On controls and Fig. 6(d)] was seen when germinated f.sp hordei conidia in locations distant from colonies, very few germlings were located in the outer zone of f. sp. avenae colonies. developed abnormally (57 %). Within the outer zone These data indicated that f. sp. hordei was more sensitive 218 T. L. W. Carver et al.

100 (a) (b) c 80 c

60

40

b 20 b b a ab a 0

100 (c) (d) c 80 c

60 b 40 Germinated conidia developing abnormally (%)

20 b a a a a 0 zone zone zone zone Inner Inner Outer Outer Control Control Distant Distant

Conidium location Conidium location relative to colony relative to colony

FIG. 6. Percentages of B. graminis €. spp. avenae, tritici and hordei conidia that germinated but formed abnormal germ tube structures (failed to form appressoria) 24 h after inoculation onto second-formed oat leaves of healthy controls, or onto leaves bearing B. graminis f. sp. avenae colonies derived from a ®rst inoculation 72 h earlier. All material was ®xed 24 h after second inoculation. Data are for f. sp. avenae conidia on oat cvs Maldwyn (a) and Selma (b), and for €. spp. tritici (c) and hordei (d) on cv. Maldwyn. On controls ( ), data are based on counts of 50 randomly selected germlings, while on leaves bearing colonies, counts were made for 50 germlings located distant from colonies ( ), 50 within the outer zone of colonies ( ),and 50 within the inner zone of colonies ( ), on each of four replicate leaves in all cases. Percentage data were transformed to angles for analysis of variance. Where columns are headed by di€erent letters, LSD tests applied to angle-transformed data indicated a signi®cant di€erence (P 5 0.05) between means within a data set.

to the in¯uences of f. sp. avenae colonies than either f. sp. There were no di€erences in the percentages of f. sp. avenae itself or f. sp. tritici. avenae germlings forming haustoria in epidermal cells of In addition to being more sensitive to the in¯uence of f. control leaves and cells distant from established colonies sp. avenae colonies, f. sp. hordei germlings appeared to show of f. sp. avenae (Fig. 8). On Maldwyn [Fig. 8(a)], there a greater range of morphological abnormalities (Fig. 7) was no di€erence between germlings distant from colonies than germlings of the other €. spp. and those within the outer zone of colonies, although within the inner zone of colonies signi®cantly fewer Haustorium formation. Germlings of f. sp. hordei never germlings formed haustoria than in any other location. In formed haustoria in oat cv. Maldwyn irrespective of the experiment using Selma [Fig. 8(b)], similar percen- whether they attacked control leaves or leaves previously tages of germlings formed haustoria in controls, in cells infected with f. sp. avenae. Haustorium formation by f. sp. distant from colonies and in the inner zone of colonies. tritici was also extremely rare; no germlings formed However, when germlings were located within the outer haustoria in control leaves or on cells distant from zone of colonies, a signi®cantly higher percentage formed established f. sp. avenae colonies. Within the outer zone haustoria than in any other location. of f. sp. avenae colonies, only 1 % of f. sp. tritici germlings formed a haustorium and within the inner zone, only Barley cv. Pallas inoculated ®rst with f. sp. hordei and 1.5 % did so. secondly with €. spp. hordei, avenae or tritici. Inhibition of B. graminis development 219

(a) (c) PGT

Hy Cu Hy

PGT Cu 30 µm 30 µm

(b) (d) Hy

PGT

Hy 30 µm 30 µm PGT

FIG. 7. Light micrographs showing examples of abnormal germling structures formed by B. graminis f. sp. hordei when conidia germinated within the boundary of f. sp. avenae colonies growing on Maldwyn oat leaves. Colonies were established from a ®rst inoculation applied 72 h earlier than the second inoculation from which the germlings derive. All material was ®xed 24 h after the second inoculation. (a) The elongated germ tube is a simple, hypha-like structure showing little sign of swelling or apical hooking, and two septa are evident. The ®rst is close to the spore, at the position where the single septum forms in morphologically normal appressorial germ tubes, and the second is approximately 35 mm from the conidium i.e. approximately where the apical appressorial hook is formed by a normal appressorial germ tube. Note: three ungerminated conidia are located close to hyphae (top right of micrograph). (b) As in (a), two septa are present in the elongated germ tube. The central cell of this tube is swollen, but a further hypha-like cell has been formed. Small protruberances are evident on the central cell, the most distal of which (directed downwards) is reminiscent of a poorly developed appressorial lobe. (c) Again two septa are present in the elongated germ tube. The central cell is swollen and shows distinct protruberances, but these are clearly too small to be mistaken for appressorial lobes. The distal cell of the germ tube is slightly swollen with a club-like apex. (d) When ®rst seen, this germling was mistakenly thought to have formed a normal appressorium with a hooked apical lobe (double arrowhead). The branch cell, separated from the hooked cell by a septum and growing left from the central portion of the hooked cell, was thought to be secondary hyphal growth. However, careful examination showed that no haustorium was present beneath the hooked structure. Therefore the cell growing left represented abnormal branching of the elongated germ tube and the hooked cell cannot be interpreted as a normal appressorial structure. Several examples of similar germlings were seen. Cu ˆ ungerminated conidium; Hy ˆ hypha of established colony; PGT ˆ primary germ tube; arrowheads indicate septa in elongated germ tube; double arrowheads indicate appressorium-like structure.

Germination. Data from barley showed trends very halved; Fig. 5). The maximum reduction was seen with f. similar to those from oat (Fig. 9). Conidia of all €. spp. sp. avenae where on distant cells approx. 69 % germinated germinated well on Pallas barley control leaves, and compared to approx. 39 % within the inner zone of f. sp. equally well on cells distant from established colonies of f. hordei colonies (i.e. a reduction of 43 %) while there were sp. hordei. There was a slight tendency for germination to only 32 and 31 % reductions in germination of f. sp. hordei be reduced when conidia were located within the outer and tritici, respectively. zone of f. sp. hordei colonies, but the reduction was small and signi®cant only in the case of f. sp. avenae. In all cases, Germling development. Extremely low frequencies of germination was signi®cantly reduced where conidia were germling abnormality were seen when conidia of any €. located within the inner zone of colonies. However, the spp. germinated within f. sp. hordei colonies growing on degree of reduction in germination was relatively smaller barley (Fig. 10). This was in complete contrast to the than that seen within the inner zone of f. sp. avenae situation seen when conidia germinated within colonies of colonies on oat (where germination rates were at least f. sp. avenae growing on oat. Even when the germlings 220 T. L. W. Carver et al.

50 a (a) (b) 40 ab c 30 b

20 a

c ab

Germinated conidia 10 forming haustoria (%) b

0 zone zone zone zone Inner Inner Outer Outer Control Control Distant Distant

Conidium location Conidium location relative to colony relative to colony

FIG. 8. Percentages of B. graminis f. sp. avenae conidia that had germinated on second-formed leaves of oat cvs Maldwyn (a) and Selma (b), and that formed morphologically normal appressoria from which plant epidermal cells were penetrated successfully, resulting in haustorium formation, 24 h after inoculation onto healthy control leaves or onto leaves bearing B. graminis f. sp. avenae colonies derived from a ®rst inoculation 72 h earlier. All material was ®xed 24 h after the second inoculation. On controls ( ), data are based on counts of 50 randomly selected germlings, while on leaves bearing colonies, counts were made for 50 germlings located distant from colonies ( ), 50 within the outer zone of colonies ( ) and 50 within the inner of colonies ( ), on each of four replicate leaves in all cases. Percentage data were transformed to angles for analysis of variance. Where columns are headed by di€erent letters, LSD tests applied to angle-transformed data indicated a signi®cant di€erence (P 5 0.05) between means within a data set. were intimately associated with hyphae of an established located within the outer zone of colonies, a few germlings colony they almost always formed normal appressoria formed haustoria (4 and 2 %, respectively). (Fig. 11). For f. sp. hordei conidia, there was no di€erence From casual observation, it was apparent that when in abnormality frequency between germlings located germlings of the inappropriate €. spp. located within a f. within the inner zone of colonies compared with controls, sp. hordei colony formed a haustorium, this haustorium germlings located distant from colonies or those located was most frequently formed in an epidermal cell that also within the colony outer zone. For €. spp. avenae and tritici, contained a haustorium of the established colony there was a signi®cant increase in the frequency of (illustrated in Fig. 13). However, data to con®rm this abnormality within the inner zone of f. sp. hordei colonies. observation were not collected. However, even here only approx. 10 % of germlings appeared abnormal (as compared with 70±80 % abnormality for any €. spp. germinating within the Double inoculation of Selma oat: hyphae of established colonies inner zone of f. sp. avenae colonies on oat). removed Removing the super®cial hyphae of B. graminis f. sp. Haustorium formation. When f. sp. hordei germlings avenae colonies that had been previously allowed to attacked control barley leaves, epidermal cells distant develop for 96 h before re-inoculation, allowed us to from established colonies, or were located within the determine whether e€ects on germination and germling outer zone of colonies, 13±14 % of appressoria were development depended upon the presence of hyphae or associated with successful penetration attempts as evi- whether the presence of haustoria within epidermal cells denced by the presence of a haustorium beneath their was sucient to mediate these e€ects. Where colony appressoria (Fig. 12). However, when germlings were hyphae were removed, haustoria formed by the colony located within the inner zone of colonies, the percentage were clearly visible, and epidermal cells containing those of appressoria with haustoria was signi®cantly increased haustoria appeared to remain alive as evidenced by lack and almost doubled in frequency (22.5 %). of whole-cell auto¯uorescence and absence of cytoplasmic On controls and on cells distant from f. sp. hordei granulation or disorganization. This was as expected colonies, germlings of €. spp. avenae and tritici never from earlier studies [8, 27±29] where haustorium- formed haustoria (Fig. 12). By contrast, when germlings containing cells also survived removal of super®cial of these inappropriate €. spp. were located within the structures. Cells containing colony haustoria obviously inner zone of f. sp. hordei colonies, they frequently formed lay within the outer boundary of colonies as de®ned a haustorium (17.5and14.5 % for avenae and tritici, earlier by an imaginary line linking tips of leader hyphae. respectively, Figs 12 and 13), and even if they were Furthermore, from our observations and Hirata's [18] Inhibition of B. graminis development 221

100 20 (a) (a) 80 15 a a a 60 b 10 40 a a a 5 a 20

0 0

100 20 (b) (b) 80 a a 15 b 60 b 10 c 40 ab 5 20 a a

Conidia that germinated (%) 0 0

100 20 (c) (c) Germinated conidia developing abnormally (%) 80 15 a a 60 ab b 10 b 40 a 5 a 20 a

0 0 zone zone zone zone Inner Inner Outer Outer Control Control Distant Distant

Conidium location Conidium location relative to colony relative to colony

FIG. 9. Percentages of B. graminis €. spp. hordei (a), avenae (b) FIG. 10. Percentages of B. graminis €. spp. hordei (a), avenae (b) and tritici (c) conidia that had germinated 24 h after inoculation and tritici (c) conidia that germinated but formed abnormal onto second-formed cv. Pallas barley leaves of healthy controls, germ tube structures (failed to form appressoria) 24 h after or onto leaves bearing B. graminis f. sp. hordei colonies derived inoculation onto second-formed cv. Pallas barley leaves of from a ®rst inoculation 72 h earlier. All material was ®xed 24 h healthy controls, or onto leaves bearing B. graminis f. sp. hordei after the second inoculation. On controls ( ), data are based on colonies derived from a ®rst inoculation 72 h earlier. All counts of 100 randomly selected conidia, while on leaves bearing material was ®xed 24 h after the second inoculation. On colonies, counts were made for 100 conidia located distant from controls ( ), data are based on counts of 50 randomly selected colonies ( ), 100 within in the outer zone of colonies ( ), and germlings, while on leaves bearing colonies, counts were made 100 within the inner zone of colonies ( ), on each of four for 50 germlings located distant from colonies ( ), 50 within the replicate leaves in all cases. Percentage data were transformed to outer zone of colonies ( ), and 50 within the inner zone of angles for analysis of variance. Where columns are headed by colonies ( ), on each of four replicate leaves in all cases. di€erent letters, LSD tests applied to angle-transformed data Percentage data were transformed to angles for analysis of indicated a signi®cant di€erence (P 5 0.05) between means variance. Where columns are headed by di€erent letters, LSD within a data set. tests applied to angle-transformed data indicated a signi®cant di€erence (P 5 0.05) between means within a data set. 222 T. L. W. Carver et al.

(a) (b)

PGT H

C C S

Hy App Hy C App H

PGT 20 µm20µm App H

FIG. 11. LTSEM micrographs of normal B. graminis f. sp. hordei germlings that developed when in intimate contact with hyphae of B. graminis f. sp. hordei growing on a barley leaf. The germlings grew from conidia inoculated 24 h earlier, while the colonies derived from inoculation made 96 h earlier. In (b) the conidium on the left has formed two subsidiary germ tubes, probably because it was deposited upon the hyphae of the colony and failed to perceive the underlying host leaf surface [38]. The primary germ tube of this conidium is presumably hidden from view. App ˆ appressorial germ tube; C ˆ conidium; H ˆ hooked apical lobe of appressorium; Hy ˆ hyphae of colony; PGT ˆ primary germ tube; S ˆ subsidiary germ tube. detailed studies, it is most likely that haustorium- Thus, in complete contrast to experiments where containing cells lay within the inner zone of colonies colonies were undisturbed, the location of conidia on whose hyphae had been removed. On the other hand, epidermal cells that had been within the boundaries of epidermal cells four cells distant from haustorium- colonies established by the ®rst inoculum had no e€ect on containing cells probably lay outside (for acropetal and either their germination or on subsequent germling basipetal cells), or only just within (for lateral cells), the development. boundaries of colonies that had been removed. Thus, these epidermal cells were regarded as comparable to Haustorium formation. Although location had no e€ect on cells previously de®ned as distant from colonies. appressorium formation, it had a dramatic e€ect on the percentage of germlings that penetrated epidermal cells Germination and germling development. Overall, slightly but successfully to produce haustoria (Fig. 14). When f. sp. signi®cantly (P 5 0.05) fewer conidia germinated on avenae germlings attacked control leaves or epidermal cells control leaves (approx. 63 %) than on double inoculated distant from cells containing a haustorium formed earlier leaves. However, unlike experiments where colony by an f. sp. avenae colony, approx. 42 % formed haustoria. hyphae were left in place, there was no di€erence in However, when attack was on a cell containing a germination between conidia distant from cells contain- haustorium formed earlier, 92 % did so. Thus, cells that ing colony haustoria (germination ˆ approx. 68 %), and contained colony haustoria were conditioned to a high conidia overlying epidermal cells containing colony level of ``accessibility'' to subsequent attack by f. sp. avenae. haustoria (germination ˆ approx. 69 %). There was a The presence of f. sp. avenae colony haustoria in signi®cant (P 5 0.001) di€erence in germination epidermal cells of oat also conditioned them to a high between the three fungal €. spp., with f. sp. hordei conidia level of accessibility to subsequent attack by germlings of germinating slightly less well (approx. 62 %) than either €. spp. tritici and hordei. Thus, 90 % of f. sp. tritici and €. spp. avenae (approx. 70 %) or tritici (approx. 68 %). 58 % of f. sp. hordei germlings penetrated successfully and These small di€erences presumably re¯ect uncontrolled formed haustoria when attacked cells contained f. sp. variation in inoculum quality between €. spp. However, avenae colony haustoria. By contrast, neither of these there was no interaction between conidium location and inappropriate €. spp. formed haustoria in control oat €. spp. indicating consistency of these small e€ects within leaves or in cells distant from cells containing f. sp. avenae the experiment. colony haustoria. For germinated conidia, there was only a very low frequency of germling abnormality in all locations and most germlings formed apparently normal appressoria. DISCUSSION For all €. spp. and all locations, abnormality was never seen for more than 6 % of germlings even if the germlings We were very surprised by our initial ®nding from Selma were in contact with epidermal cells containing haustoria oat that germination and germling development were formed by f. sp. avenae colonies that had been removed. markedly and adversely a€ected when B. graminis conidia Inhibition of B. graminis development 223

50 (a) Ct 40 Hat Appt Appt Ct Hat 30 b 20 a a a 10 Hh 0 Apph Ch 50 (b) 40 FIG. 13. Light micrograph of two B. graminis f. sp. tritici germlings located within the inner zone of a B. graminis f. sp. 30 hordei colony. The colony was established from a f. sp. hordei conidium deposited 72 h earlier than the f. sp. tritici conidia. 20 b The specimen was ®xed 24 h after deposition of the f. sp. tritici conidia. Both f. sp. tritici germlings produced normal appres- 10 soria from which haustoria were formed. Apph ˆ primary a appressorium of f. sp. hordei colony; Appt ˆ appressorium of f. NA NA 0 sp. tritici germling; Ch ˆ conidium of f. sp. hordei,Ct ˆ conidium of f. sp. tritici.Hh ˆ primary haustorium of f. sp. hordei colony; 50

Germinated conidia forming haustoria (%) Hat ˆ primary haustorium of f. sp. tritici germling. (c) 40 were deposited within established oat mildew colonies. 30 This was because a considerable body of evidence indicates that deposition of B. graminis conidia, and 20 conidia of other powdery mildew fungi, close to an b established B. graminis colony can greatly enhance the 10 chances of these conidia producing successful infections. a NA NA This evidence comes from investigations of wheat and 0 barley infected with their appropriate f. sp. and sub- sequently inoculated with an inappropriate f. sp. of zone zone Inner Outer B. graminis or with less closely related Erysiphaceae. Thus Control Distant Moseman and his co-workers [30, 31] found that wheat Conidium location could be attacked by B. graminis f. sp. hordei if leaves were relative to colony already infected with a virulent isolate of B. graminis f. sp. tritici or if the two fungi were inoculated simultaneously. FIG. 12. Percentages of B. graminis €. spp. hordei (a), avenae (b) and tritici (c) conidia that had germinated on second-formed cv. Tsuchiya and Hirata [37] inoculated B. graminis f. sp. Pallas barley leaves, and that formed morphologically normal hordei-infected barley with f. sp. tritici,aB. graminis isolate appressoria from which plant epidermal cells were penetrated from Agropyron, or with powdery mildew fungi from 49 successfully, resulting in haustorium formation, 24 h after di€erent dicotyledonous species; the fungi included inoculation onto healthy control leaves or onto leaves bearing species from the genera Erysiphe, Sphaerotheca, Podosphaera, B. graminis f. sp. hordei colonies derived from a ®rst inoculation Microsphaera and Uncinula. Of the 51 fungi tested, 45 were 72 h earlier. All material was ®xed 24 h after the second inoculation. On controls ( ), data are based on counts of 50 able to infect the barley and 30 formed sporulating randomly selected germlings, while on leaves bearing colonies, colonies. Tsuchiya and Hirata [37] noted that these counts were made for 50 germlings located distant from colonies infections almost invariably occurred close to established f. ( ), 50 within the outer zone of colonies ( ) and 50 within the sp. hordei colonies and detailed histological studies of the inner zone of colonies ( ), on each of four replicate leaves in all B. graminis isolate from Agropyron and of Sphaerotheca cases. Percentage data were transformed to angles for analysis of fuliginea showed that successful infections by these fungi variance. Where columns are headed by di€erent letters, LSD tests applied to angle-transformed data indicated a signi®cant were almost invariably established in barley epidermal di€erence (P 5 0.05) between means within a data set. NA: cells containing B. graminis f. sp. hordei haustoria or in data ˆ zero, therefore excluded from analyses. neighbouring cells. Similarly, Ouchi et al.[33] concluded that induced `accessibility' of barley to attack by wheat 224 T. L. W. Carver et al.

B. graminis f.sp.

avenae tritici hordei 100

80

60

40

20 Germinated conidia forming haustoria (%) 0 Control Distant Overlying Overlying Overlying

Location of appressorium in relation to epidermal cells containing haustoria formed by colony

FIG. 14. Percentages of B. graminis f. sp. avenae, tritici and hordei germlings that formed appressoria and penetrated Selma oat epidermal cells to form a haustorium 36 h after inoculation. Data are for germlings on healthy control leaves, or on leaves that had supported f. sp. avenae colonies derived from a ®rst inoculation applied 96 h before test conidia were inoculated, but from which all super®cial fungal structures (conidia, hyphae) were removed immediately before inoculation with test conidia. On controls ( ) data are based on counts of 50 randomly selected germlings, while on leaves that had supported colonies data are based on counts of 25 germlings whose appressoria were on epidermal cells that were distant from cells containing haustoria formed earlier by colonies ( ), and 25 germlings with appressoria overlying cells containing haustoria formed earlier by colonies (h), on each of four replicate leaves in each case. Data for €. spp. tritici and hordei on healthy oat control leaves and on epidermal cells distant from haustorium- containing epidermal cells were zero in all cases (not shown). mildew and S. fuliginea was a phenomenon expressed majority of germinated conidia di€erentiated morpho- locally in the immediate vicinity of established barley logically normal appressoria. The functionality of these mildew colonies. More recent and detailed studies by appressoria was demonstrated by the fact that a higher Kunoh and his co-workers [22, 24, 25] again showed that percentage were associated with successful penetration induced accessibility of barley coleoptile cells to attack by and haustorium formation than where conidia were the non pathogen Erysiphe pisi, is an extremely localized remote from established colonies. Indeed, our ®ndings e€ect dependent on prior or near simultaneous infection were comparable to earlier studies of powdery mildewed by B. graminis f. sp. hordei. Thus, many studies have shown barley and wheat [30, 31, 37] in showing that inappro- that barley and wheat can be rendered susceptible to priate €. spp., normally unable to infect healthy barley, powdery mildew fungi that are normally unable to infect frequently formed haustoria if they attacked within the them, if they are previously infected by a pathogenic inner zone of barley powdery mildew colonies. The ability powdery mildew fungus. However, this depends upon the of the inappropriate €. spp. to infect barley, and increased normally non-pathogenic fungus being in close proximity penetration frequency by B. graminis f. sp. hordei,was to established infections of the pathogenic B. graminis. This largely lost if attacks were made within the outer zone of in turn clearly suggests that proximity to an established colonies or if they were distant from them. These B. graminis colony does not greatly impede germination or observations support the conclusion from earlier histo- development of later-applied propagules and in fact logical studies that prior infection of barley by a favours their ability to infect. Hence, we were surprised compatible isolate decreases epidermal cells' ability to to ®nd that this was not the case when B. graminis conidia resist attack [27±29]. However, also in agreement with were deposited within oat mildew colonies. these studies, this e€ect appears to be localized to cells Nevertheless, our observations of the behaviour of three containing pre-formed haustoria and to neighbouring di€erent B. graminis €. spp. deposited within the inner cells. zone of established barley mildew colonies were broadly in When conidia of all three €. spp. were deposited within line with expectation. Although germination by conidia of established oat mildew colonies germination percentages all three €. spp. was somewhat depressed within these were depressed. This was similar to, but even more colonies, and the frequency of abnormal long germ tubes marked than the situation seen within barley mildew was slightly increased within the inner colony zone, the colonies. We know from previous work [8] that the Inhibition of B. graminis development 225 presence of a haustorium within an oat epidermal cell within the colony inner zone. The inner zone recognized should have induced accessibility to subsequent attack by at the time of ®xation probably represented (approxi- B. graminis f. sp. avenae. However, we could not be sure mately) the limits of hyphal development extant at the from our observations of conidia deposited within time when second inoculations were made (Fig. 2) established oat mildew colonies whether the plant cells suggesting that conidia were close to, or in contact had been a€ected. This was because very few germinated with, hyphae from the time of their deposition. It is not conidia produced morphologically normal appressoria, clear however, whether spores that failed to germinate and even where appressorial germ tubes formed an were actively and rapidly killed by factors released by the apparently normal, hooked apical lobe, we could not be colony or whether they survived for some time (perhaps sure if the structure was functional. Thus, very few until ®xation) although processes leading to germ tube germlings were capable of attempting infection and this emergence were inhibited. Further studies should be able alone could explain the very low frequency of haustorium to distinguish these possibilities and in turn shed light on formation seen within oat mildew colonies. the likelihood that similar factor(s) suppress the germina- Nevertheless, there was some evidence from Selma oat tion of conidia attached to the spore chain which (Fig. 8) that accessibility to B. graminis f. sp. avenae may obviously can remain alive for some time before release. have been induced within the outer zone of established Abnormal germling development was extremely fre- colonies, and some evidence of induced accessibility to f. quent when conidia of all B. graminis €. spp. germinated sp. tritici within oat mildew colonies established on within oat mildew colonies. By comparison, abnormal- Maldwyn oat. However, clear evidence for induced ities were rarely seen within barley mildew colonies, even accessibility in oat epidermal cells was provided only where germlings of the second inoculum were in very when we removed the super®cial structures of the oat close contact with hyphae (Fig. 11). The latter ®nding mildew colonies prior to applying the second inoculum. appears to contradict Hirata's [18] observations of the Thus, when germlings of all three €. spp. attacked barley mildew system. Hirata [18] stated that . . . ``when epidermal cells containing a haustorium formed earlier (barley) leaves bearing old (barley) mildew colonies are by a oat mildew colony, a substantial proportion were inoculated (with B. graminis f. sp. hordei conidia), conidia successful in penetrating the cells and establishing a that fall on or near the peripheral part of the colonies functional haustorium. For f. sp. avenae, this was a germinate and develop long septate germ tubes. The quantitative e€ect where the frequency of successful germ tube does not generally form an appressorium and penetration was increased relative to control leaves or the primary haustorium''. Hirata's drawings of these epidermal cells distant from haustorium-containing cells. germ tubes resemble the abnormal germlings we saw For €. spp. tritici and hordei the e€ect was qualitative in within the inner zone of oat mildew colonies. However, the sense that these fungi were virtually unable to infect Hirata did not quantify the frequency of this occurrence, oat cells that did not contain a f. sp. avenae haustorium. and he did not indicate exactly the age of the colonies The available evidence clearly shows that prior involved in his observations. It is possible that the ability infection of cereal epidermal cells by B. graminis has of powdery mildew colonies to generate/release chemicals dramatic, albeit localized, consequences for the ability of that cause abnormal germling development is a function cells to resist subsequent attack by the fungus. These of their age. Thus, although our oat mildew colonies had e€ects a€ect not only the outcome of attacks by obviously released the factor(s) within 4 days of compatible isolates of the appropriate f. sp., but extend inoculation, barley mildew colonies of this age had not to permit invasion of cells by inappropriate €. spp. and done so. If we had incubated our barley mildew colonies non pathogenic species. The basis of induced changes in for longer before we applied the second inoculum, we accessibility remain unknown. may have detected e€ects on germling development. At present, we have no idea what causes suppression of Hirata [18] also found that anastomosis occasionally germination by conidia deposited within established occurred when abnormal long germ tubes encountered powdery mildew colonies. However, it seems that similar hyphae. We never saw this. He went on to show that such factors operate in colonies of both barley and oat powdery anastomosis could occur between conidial germ tubes of mildew, and the factors were e€ective against all three €. powdery mildew from Agropyron (a non-pathogen of spp. that we tested. It is a truism that powdery mildew barley) and hyphae of established barley mildew colonies. conidia do not germinate while they are within From Tsuchiya and Hirata's [37] further discussion of the developing spore chain attached to the mother these studies, it appears that following anastomosis with conidiophore, but that they germinate rapidly after barley powdery mildew, the powdery mildew from release. It seems possible that the same factor(s) suppress Agropyron derived nutrients from the barley mildew germination by conidia deposited within an established colony. This allowed the Agropyron fungus to develop colony. The factor(s) are not particularly mobile as the hyphae, conidiophores and conidia capable of re- suppressive e€ect was usually greater when conidia lay infecting Agropyron. Clearly, Hirata and his co-workers' 226 T. L. W. Carver et al. studies have important implications with respect to the plant's physiological or biochemical status. Thus, when intimacy of genetic and physiologic association resulting super®cial hyphae were removed from oat leaves, spores from anastomoses formed between fungal individuals deposited on haustorium-containing cells germinated and both within and between €. spp. of B. graminis and normal, functional appressoria developed with at least possibly between genera of the Erysiphales. There is the same frequency as on healthy leaves. The induced immense scope to pursue these studies. accessibility of haustorium-containing cells showed that Presumably the e€ective factor(s) produced by oat they remained radically altered, at least in terms of their mildew colonies that caused abnormal germling devel- accessibility, by the presence of pre-formed haustoria. If opment in our experiments must be di€erent from those factors inhibitory to germination and germling develop- that suppress germination. This conclusion arises from ment were generated within haustorium-containing cells, our ®nding that germination frequency was consistently it seems highly likely that removal of the epicuticular wax reduced within colonies of both f. sp. hordei and f. sp. layer (that inevitably occurs during cellulose acetate avenae, whereas our barley powdery mildew colonies had stripping [7, 10, 11, 27]) would have facilitated their little if any detectable e€ect on germling development. movement to the cell surface and increased inhibitory Again, however, we have no idea what factor(s) may e€ects. This was clearly not so. These ®ndings again cause abnormal germling development. We believe the appear to contradict Hirata's [18] observations. Hirata factor(s) has limited mobility, because, as with suppres- found that abnormal germlings often developed when sion of germination, greater e€ect was evident within the barley mildew conidia were inoculated onto a leaf from inner than the outer colony zone. Nevertheless, direct which older colonies had previously been ``scraped'' contact with a hypha was not necessary for abnormal away. We assume that ``scraping'' involved a physical germling development (Figs 4 and 7). Therefore, the procedure that could have ruptured the cell walls of factor(s) had some ability for movement over the leaf super®cial fungal structures (hyphae, conidiophores, etc.) surface. Where conidia germinated, the factor(s) did not releasing their cell contents onto the leaf surface prior to appear to inhibit PGT function since germlings often application of test conidia. If the factor(s) that cause developed subsequent germ tubes that elongated sub- germling abnormality are generated within cells of the stantially. Such elongation occurs only if PGTs function colony, these may well have been released to remain as correctly in recognising inductive features of the plant residue on the leaf surface following ``scraping'' in surface [4±7, 38]. The factor's major e€ect was to inhibit Hirata's experiment. Our technique would have removed the later stages of appressorial germ tube di€erentiation. all such residues along with the leaf epicuticular wax. We Thus, elongating germ tubes remained hypha-like, failed suggest this explains the apparent di€erence between our to swell and di€erentiate a hooked apical lobe, and often observations and Hirata's. developed peculiar branches and multiple septa. Such Our current work indicates that when B. graminis abnormalities could result from failure of the elongating conidia are deposited within an established colony their tube itself to recognize plant surface features [4±7]. This germination is suppressed. Our experiments and Hirata's seems unlikely, however, because the e€ective factor(s) seminal studies show that where conidia do germinate did not a€ect the PGT's ability to recognize such within a colony, their subsequent development can be features. Alternatively, it may be that the oat mildew disrupted so that they do not form appressoria or colonies release factor(s) that block intracellular signal- haustoria, and are therefore incapable of independent ling or signal transduction pathways that regulate existence and reproduction. Why should established appressorial formation. Temporal regulation of such colonies produce factors that inhibit germination and pathways is known to be associated with maturation disrupt germling development? In situations with geneti- and hooking of the appressorial germ tube; pharmaco- cally diverse host and pathogen populations, there could logical inhibitors of these pathways can prevent appres- be obvious selective advantages for an established colony sorium formation and lead to the development of in preventing potential competitors from infecting nearby undi€erentiated, hypha-like elongated germ tubes [16, plant epidermal cells which have been rendered highly 17, 19]. Perhaps, our oat powdery mildew colonies accessible to penetration by the presence of haustoria produced chemicals with similar e€ects. Whatever the formed by the established colony. These competitors may explanation, it is intriguing that the e€ective factor(s) be less-well adapted to parasitize a particular plant disrupt germling development but are presumably genotype, and may even be avirulent. If this were so, harmless to the established colony. avirulent individuals could kill attacked cells and deprive Our experiments suggest strongly that the factors the adapted colony of potential feeding sites. On the other suppressing B. graminis germination and causing abnor- hand, some conidia produced by a colony are likely to be mal germling development were directly associated with deposited close to that colony; suppressing their develop- presence of oat mildew hyphae, and are not due to ment would prevent competition for nutrients in ®nite indirect e€ects mediated via modi®cation of the host oat local supply within a niche already occupied by that Inhibition of B. graminis development 227 pathogen genotype. Similar e€ects are likely to apply 7. Carver TLW, Kunoh H, Thomas BJ, Nicholson RL. within agricultural crops although their selective 1999. Release and visualization of the extracellular matrix of conidia of Blumeria graminis. Mycological Research 103: advantage may be less marked. According to Hirata's 547±560. [18] observations, conidia that germinate and form 8. Carver TLW, Lyngkjñr MF, Neyron L, Strudwicke CC. abnormal germ tubes at the periphery of an established 1999. Induction of cellular accessibility and inaccessibility colony, may be able to form anastomoses with hyphae of and suppression and potentiation of cell death in oat that colony and, in e€ect, parasitize the colony by using its attacked by Blumeria graminis f. sp. avenae. Physiological and resources. It remains to be determined how frequently this Molecular Plant Pathology 55: 183±196. 9. Carver TLW, Phillips M. 1982. E€ects of photoperiod and happens, but it seems unlikely that the frequency of level of irradiance on production of haustoria by Erysiphe anastomosis compensates in any signi®cant way for the graminis f. sp. hordei. Transactions of the British Mycological suppressive e€ects established colonies have on germina- Society 79: 207±211. tion and appressorium formation by potential competi- 10. Carver TLW, Thomas BJ. 1990. Normal germling devel- tors. opment by Erysiphe graminis on cereal leaves freed of epicuticular wax. Plant Pathology 39: 367±375. Currently we are attempting to isolate the factors 11. Carver TLW, Thomas BJ, Ingerson-Morris SM. 1995. responsible for inhibiting spore germination and germling The surface of Erysiphe graminis and the production of development within oat mildew colonies. It is conceivable extracellular material at the fungus-host interface during that this may reveal novel means of disease control. germling and colony development. Canadian Journal of Botany 73: 272±287. 12. Carver TLW, Williams O. 1980. The in¯uence of photoperiod on growth patterns of Erysiphe graminis f. sp. We formally acknowledge the pioneering research con- hordei. Annals of Applied Biology 94: 405±414. tribution made by Professor K. Hirata (latterly Amano) 13. Carver TLW, Zeyen RJ, Robbins MP, Dearne GA. 1992. that laid the foundations for the studies reported here, E€ects of the PAL inhibitor, AOPP, on oat, barley, and and his many other painstaking works that have wheat cell responses to appropriate and inappropriate contributed so much to our current understanding of formae speciales of Erysiphe graminis DC. Physiological and Molecular Plant Pathology 41: 397±409. powdery mildew biology. The research reported here was 14. Francis S, Dewey M, Gurr SJ. 1996. The role of cutinase supported by MAFF (U.K.) under Project CE0154 and in germling development in Erysiphe graminis. Physiological in part by the Danish Agricultural and Veterinary and Molecular Plant Pathology 49: 201±211. Research Council. The Institute of Grassland and 15. Genstat 5: Procedure Library Manual Release PL 10. Environmental Research is supported by the Biotechnol- 1997. Payne RW, Morgan GW and Arnold GM, eds. ogy and Biological Sciences Research Council. Oxford, U.K.: Numerical Algorithms Group, 371. 16. Hall AA, Bindslev L, Rouster J, Rasmussen SW, Oliver R, Gurr SJ. 1999. The involvement of cAMP and PKA in appressorial di€erentiation in Erysiphe graminis f. sp. hordei. Molecular Plant Microbe Interactions 12: 960±968. 17. Hall AA, Gurr SJ. 2000. Initiation of appressorial germ REFERENCES tube di€erentiation and appressorial hooking: distinct 1. Aist JR, Bushnell WR. 1991. Invasion of plants by morphological events regulated by cAMP-signalling in powdery mildew fungi, and cellular mechanisms of Blumeria graminis f. sp. hordei. Physiological and Molecular resistance. In: Cole GT, Hoch HC, eds. The Spore and Plant Pathology 56: 39±46. Disease Initiation in Plants and Animals. New York, U.S.A.: 18. Hirata K. 1967. Notes on haustoria, hyphae and conidia of Plenum Press, 321±345. the powdery mildew fungus of barley. Memoirs of the 2. Carver TLW, Bushnell WR. 1983. The probable role of Faculty of Agriculture, Niigata University 6: 207±259. primary germ tubes in water uptake before infection by 19. Kinane JS, Dalvin S, Bindslev L, Hall AA, Gurr SJ, Erysiphe graminis. Physiological Plant Pathology 23: 229±240. Oliver R. 2000. Evidence that cAMP ¯uxes control early 3. Carver TLW, Carr AJH. 1978. The early stages of mildew development of barley mildew. Molecular Plant±Microbe colony development on susceptible oats. Annals of Applied Interactions 13: 494±502. Biology 89: 201±209. 20. Kunoh H, Hayashimoto A, Harui M, Ishizaki H. 1985. 4. Carver TLW, Ingerson SM. 1987. Responses of Erysiphe Induced susceptibility and enhanced resistance at the graminis germlings to contact with arti®cial and host cellular level in barley coleoptiles. I. The signi®cance of surfaces. Physiological and Molecular Plant Pathology 30: timing of fungal invasion. Physiological Plant Pathology 27: 359±372. 43±54. 5. Carver TLW, Ingerson SM, Thomas BJ. 1996. In¯uences 21. Kunoh H, Katsuragawa N, Yamaoka N, Hayashimoto of host surface features on development of Erysiphe graminis A. 1988. Induced accessibility and enhanced inaccessi- and Erysiphe pisi. In: Kerstiens G, ed.. Plant CuticlesÐAn bility at the cellular level in barley coleoptiles. III. Timing Integrated Functional Approach. Oxford, U.K.: Bios Scienti®c and localization of enhanced inaccessibility in a single Publishers, 255±266. coleoptile cell and its transfer to an adjacent cell. 6. Carver TLW, Ingerson-Morris SM, Thomas BJ, Zeyen Physiological and Molecular Plant Pathology 33: 81±93. RJ. 1995. Early interactions during powdery mildew 22. Kunoh H, Komura T, Kobayashi I, Yamaoka N. 1990. infection. Canadian Journal of Botany 73(suppl. 1): Induced accessibility and enhanced inaccessibility at the S632±S639. cellular level in barley coleoptiles. VII. Enhancement of 228 T. L. W. Carver et al. inaccessibility at the prepenetration stage of a nonpatho- 30. Moseman JG, Greeley LW. 1964. Predisposition of wheat gen. Physiological and Molecular Plant Pathology 37: 399±407. by Erysiphe graminis f. sp. tritici to infection by Erysiphe 23. Kunoh H, Kuroda K, Hayashimoto A, Ishizaki H. 1986. graminis f. sp. hordei. Phytopathology 54: 618. Induced susceptibility and enhanced resistance at the 31. Moseman JG, Scharen AL, Greeley LW. 1965. Propa- cellular level in barley coleoptiles. II. Timing and location gation of Erysiphe graminis f. sp. tritici on barley and Erysiphe of induced susceptibility in a single coleoptile cell and its graminis f. sp. hordei on wheat. Phytopathology 55: 92±96. transfer to an adjacent cell. Canadian Journal of Botany 64: 32. Nielsen KA, Nicholson RL, Carver TLW, Kunoh H, 889±895. Oliver RP. 2000. First Touch: An immediate response to 24. Kunoh H, Kuroda K, Toyoda K, Yamaoka N, Kobaya- surface recognition in conidia of Blumeria graminis. shi I. 1991. Induced accessibility and enhanced inaccessi- Physiological and Molecular Plant Pathology 56: 63±70. bility at the cellular level in barley coleoptiles. IX. 33. Ouchi S, Oku H, Hibino C. 1976. Localization of induced Accessibility induced by Erysiphe graminis which promotes resistance and susceptibility in barley leaves inoculated the subsequent infection of challenging E. graminis. Annals with the powdery mildew fungus. Phytopathology 66: of the Phytopathological Society of Japan 57: 57±60. 901±905. 25. Kunoh H, Toyoda K, Yamaoka N, Kobayashi I. 1989. 34. Ouchi S, Oku H, Hibino C, Akiyama I. 1974. Induction Induced accessibility and enhanced inaccessibility at the of accessibility and resistance in leaves of barley by some races of Erysiphe graminis. Phytopathologische Zeitschrift 79: cellular level in barley coleoptiles. V. Duration of stimulus 24±34. by a non-pathogen in relation to enhanced inaccessibility. 35. Pryce-Jones E, Carver TLW, Gurr SJ. 1999. The roles of Physiological and Molecular Plant Pathology 35: 507±518. cellulase enzymes and mechanical force in host penetra- 26. Kunoh H, Tsuzuki T, Ishizaki H. 1978. Cytological tion by Erysiphe graminis f. sp. hordei. Physiological and studies of early stages of powdery mildew in barley and Molecular Plant Pathology 55: 175±182. wheat leaves. IV. Direct ingress from super®cial primary 36. Suzuki S, Komiya Y, Mitsui T, Tsuyumu S, Kunoh H, germ tubes and appressoria of Erysiphe graminis hordei on Carver TLW, Nicholson RL. 1998. Release of cell wall barley leaves. Physiological Plant Pathology 13: 327±333. degrading enzymes from conidia of Blumeria graminis on 27. Lyngkjñr MF, Carver TLW. 1999a. Induced accessibility arti®cial substrata. Annals of the Phytopathological Society of and inaccessibility to Blumeria graminis f. sp. hordei in barley Japan 64: 160±167. epidermal cells attacked by a compatible isolate. Physio- 37. Tsuchiya K, Hirata K. 1973. Growth of various powdery logical and Molecular Plant Pathology 55: 151±162. mildew fungi on barley leaves infected preliminarily with 28. Lyngkjñr MF, Carver TLW. 1999b. Modi®cation of mlo5 the barley mildew fugus. Annals of the Phytopathological resistance to Blumeria graminis attack in barley as a Society of Japan 39: 399±403. consequence of induced accessibility and inaccessibility. 38. Wright AJ, Carver TLW, Thomas BJ, Fenwick NID, Physiological and Molecular Plant Pathology 55: 163±174. Kunoh H, Nicholson RL. 2000. The rapid and accurate 29. Lyngkjñr MF, Carver TLW. 2000. Conditioning of cereal determination of germ tube emergence site by Blumeria leaf cellular defence responses to powdery mildew by prior graminis conidia. Physiological and Molecular Plant Pathology attack. Molecular Plant Pathology 1: 41±50. 57: 281±301.