Fungal suppression of resistance against inappropriate Blumeria graminis formae speciales in barley, oat and wheat

Abstract

When barley, wheat or oat leaf epidermal cells were attacked by their appropriate forma specialis (f.sp.) of Blumeria graminis DC. Speer (f.sp. hordei, tritici and avenae, respectively), many attempted penetrations succeeded, functional haustoria were formed and very few plant cells died. When attacked by either of the two possible inappropriate ff.spp., penetration attempts failed in association with papilla deposition by epidermal cells, attacked cells died, or if visible haustoria were formed the plant cell died very soon afterwards. Double inoculation experiments were performed where each cereal species was first attacked by its appropriate f.sp., as inducer, and later by the different ff.spp. as challenger. Infection by the appropriate inducer profoundly affected cellular responses to challenger attack. Suppression of defensive responses was dramatic within epidermal cells containing the inducer haustorium, evident to some extent in adjacent cells, but undetectable at two cells distance. Suppression of penetration resistance allowed most challenger attacks, even by inappropriate ff.spp., to form a haustorium. Furthermore, death of penetrated epidermal cells was also suppressed so that haustoria of the inappropriate ff.spp. functioned to support colony development. In oat, delayed epidermal cell death prevented full colony development by inappropriate ff.spp., but in barley and wheat, no cell death was apparent by four days after inoculation and colonies of the inappropriate ff.spp. produced extensive hyphae, secondary haustoria and conidial chains.

Introduction

The powdery mildew fungus, Blumeria graminis DC. Speer (syn. Erysiphe graminis DC) is an obligate biotrophic fungus with specialised physiologic forms (formae speciales; ff.spp.) capable of infecting specific genera of the Gramineae [20], [36], [49]. Specificity at the f.sp.—genus level is believed to follow a gene-for-gene relationship [17], [21], [47], [48], [49], [50]. Hiura [21] and Tosa [48] speculated that B. graminis ff.spp. may have evolved from an original unspecialised population that hybridised freely. It was suggested that fungal specialisation developed gradually through co-evolution with gramineous plants, leading eventually to the establishment of the ff.spp. [17], [21], [48]. If this is correct, the present ff.spp. presumably still share many loci for pathogenicity [47]. For the purposes of the current paper, we will use the term ‘appropriate’ to describe fungal f.sp./plant species combinations that normally allow full disease development in the absence of effective race specific resistance. Conversely, we will use ‘inappropriate’ to describe fungal f.sp./plant species combinations that do not normally allow full disease development.

Under favourable conditions, B. graminis conidia germinate and form an appressorium from which an infection peg emerges approx. 15 h after inoculation and attempts to penetrate the plant epidermal cell wall. These early stages of development are achieved equally successfully on appropriate and inappropriate cereal species [8], [23]. On an appropriate host where penetration succeeds, the tip of the peg swells and eventually forms a digitate haustorium that will continue to grow for 4–5 days [3], [19]. Shortly after formation of the digitate processes, approximately 20–24 h after inoculation, secondary hyphae emerge from the appressorial germ tube and/or mother conidium. These branch and elongate at an increasing rate over the next 2–3 days, before hyphal appressoria form as lateral protrusions from the hyphal cells. Penetration pegs from these hyphal appressoria then attempt to penetrate, and if successful, second generation haustoria are formed [3], [19]. Subsequent generations of appressoria and haustoria are formed in a circadian cycle each subsequent day [3], [19]. Conidiophores are first formed after approximately 4–5 days and conidial chains shortly afterwards.

Not all attacks by functional appressoria of an appropriate f.sp. succeed. For a proportion of attacks, the infection peg fails to penetrate the host cell. For many years it has been thought that papillae (chemically complex appositions deposited by living plant cells [58]) are involved in this penetration resistance [1], [58]. Papilla deposition is also thought to be a major factor conferring resistance of cereals to inappropriate ff.spp. of B. graminis [8], [12], [13], [22], [23], [45], [59]. In appropriate combinations where the host lacks race-specific resistance, epidermal cells that are penetrated usually survive to support haustorial development and biotrophy. However, a small percentage of attacked cells die even in such compatible interactions [4], [8]. In inappropriate combinations, penetration almost invariably leads to cell death, commonly before the haustorium develops digitate processes [8], [23], [51].

A body of evidence shows that if an appropriate f.sp. of B. graminis penetrates and forms a functional haustorium within a cereal epidermal cell, the ability of that cell to resist penetration by a subsequent attack is almost totally suppressed [4], [30], [31], [32], [33], [35]. This is a local effect being expressed to a lesser degree in adjacent epidermal cells but generally being lost at two cells distance from the infected cell. In addition, the presence of a functional haustorium can also suppress hypersensitive cell death conditioned by race-specific resistance. Thus, when barley epidermal cells contain a haustorium of a virulent fungal isolate, subsequent attack by a normally avirulent isolate will succeed, the plant cell will live, the avirulent haustorium will develop and function, and the colony will grow [35]. Furthermore, when Tsuchiya and Hirata [52] inoculated B. graminis f.sp. hordei-infected barley with f.sp. tritici, a B. graminis isolate from Agropyron, or with powdery mildew fungi from many different dicotyledonous species, 45 of 51 tested fungi were able to infect the barley and 30 formed sporulating colonies. Histological investigation indicated that successful infections by these fungi were almost invariably established in barley epidermal cells containing B. graminis f.sp. hordei haustoria or in neighbouring cells. Similarly, more recent and detailed studies by Kunoh and his co-workers [28], [29] also showed that the ability of the non-pathogen Erysiphe pisi to infect barley coleoptile cells is localised to cells infected by B. graminis f.sp. hordei. Thus, prior infection and haustorium formation by an appropriate f.sp. of B. graminis has profound effects on epidermal cell resistance. It can induce susceptibility to infection not only by virulent and avirulent isolates of an appropriate f.sp. but also by fungal genera that are not normally pathogenic.

The current paper is focussed on detailing the effects that infection by an appropriate f.sp. may have on subsequent attack by inappropriate ff.spp. It is not known how profound these effects are or the extent to which induced changes may be transmitted between leaf epidermal cells. However, there is good circumstantial evidence indicating that prior infection by an appropriate B. graminis f.sp. can suppress resistance to subsequent attacks by inappropriate ff.spp. Thus, Hiura [20] found that when barley, wheat or Agropyron were infected by an appropriate f.sp. of B. graminis, subsequent inoculation with an inappropriate f.sp. led to cleistothecium formation. This must have resulted from hybridisation of the two ff.spp. suggesting that the inappropriate f.sp. had become established on the plant. This suggestion was supported by later work [37] showing that wheat plants infected with B. graminis f.sp. tritici could support sporulation by f.sp. hordei, and that barley plants infected with B. graminis f.sp. hordei could support sporulation by f.sp. tritici. More detailed histological studies [38], [39], [40], [41] suggested that in barley infected by B. graminis f.sp. hordei induction of susceptibility (‘accessibility’) to f.sp. tritici, was a phenomenon expressed locally in the immediate vicinity of established f.sp. hordei colonies.

Incidental observations from recent studies [6] support the conclusion that if barley or oat cells contain a functional haustorium of an appropriate B. graminis f.sp., those cells become susceptible to penetration by inappropriate ff.spp. However, this effect could not be quantified fully and no attempt was made to assess its transmission between epidermal cells. Therefore, the present work used a double inoculation procedure whereby B. graminis f.sp. hordei, tritici and avenae were allowed to form haustoria in barley, wheat and oat, respectively, before infected cells were attacked later by appropriate and inappropriate ff.spp. Details of fungal development and host responses were related to sites of initial, successful attacks.