Introduction

Some of the most serious plant pathogens world wide are obligate biotrophic parasites. This term characterizes a specific lifestyle in which the host as a whole suffers only minor damage over a longer period of time. The pathogen on the other hand depends on a living host to complete its life cycle. In order to mark off the true obligate biotrophic fungi from hemibiotrophs or necrotrophs we suggest the following criteria: a) highly differentiated infection structures; b) limited secretory activity; c) narrow contact zones separating fungal and plant plasma membranes; d) long term suppression of host defense responses; e) the formation of haustoria.

Haustoria are specifically differentiated hyphae breeching the plant cell wall and penetrating into the host cell. These structures have generated the interest of plant pathologists ever since their discovery by the Italian botanist Giovanni Zanardini about 150 years ago. The name was introduced by the German physician Heinrich Anton de Bary, and is derived from the Latin word haustor, which means the pail. This illustrates that the structure and location of these haustoria already suggested a function in nutrient uptake to researchers one and a half centuries ago. For a summary on haustorial structure and function see (Voegele and Mendgen, 2003; Mendgen and Hahn, 2002, Hahn and Mendgen, 2001).

Transmission electron micrograph image of a haustorium (Szabo & Bushnell. 2001. PNAS USA 98:7654-5;left) and schematic representation of the ultrastructural details (right).

However, knowledge about this key element of the obligate biotrophic life style is still fairly scarce. The main reasons for this are the lack of stable transformation systems for haustoria forming fungi and the fact that these organs are not formed in culture. This excludes them from the application of many molecular techniques successfully used for other systems. Claviceps purpurea, Ustilago maydis and several other hemibiotrophs for example are well-studied and amenable to molecular tools. However, studies of these organisms can only give limited evidence about the processes and types of interactions involved in a true obligate biotrophic relationship, since haustoria are not formed in these systems.

The work of our group is focused on the elucidation of the function(s) of haustoria applying a combination of molecular, biochemical and cytological techniques.

Research Projects

Nutrient Acquisition in Rust Fungi

An obvious advantage for obligate biotrophic fungi is the fact that they can exploit the biosynthetic capacities of their living hosts. Rather than going through complex pathways for the provision of nutrients they have the alternative of using a continuous supply of metabolites provided by the plant.

Our analysis of nutrient uptake in U. fabae focuses mainly on carbohydrates. We have identified, characterized and localized the so far only monosaccharide transporter, HXT1, of a rust fungus (Voegele et al., 2001).

We were able to show that transcript and gene product are exclusively localized in haustoria.

HXT1 transcripts are only found in haustoria and infected leaves. A: schematic representation of rust infection structures. B: loading control. C: Northern Blot. 1, uredospore; 2, germtube; 3, appressorium; 4, substomatal vesicle; 5, infection hyphae; 6, haustorial mother cell; 7, isolated haustoria; 8, infected leaves; 9, non-infected leaves; bAp: bulk apoplast; NB: neckband; EM: extrahaustorial matrix. The number on the right gives the size estimate in kb.		Localization of HXT1p in the periphery of fully developed haustoria (A) and along the haustorial plasma membrane (B) using S651p. A: Superimposed interference contrast and fluorescence images. h: haustorium; hn: haustorial neck; c: plant cell. Bar = 5 µm. B: Electron micrograph depicting gold labeling along the haustorial plasma membrane (hpm), but no labeling over the haustorium (h), extra haustorial matrix (ehma), extrahaustorial membrane (ehm), or plant cell (c). Bar: 0.1 µm

A biochemical description of the gene product was only possible in a heterologous expression system because of the biotrophic nature of rust fungi. Function of HXT1p was analyzed in two different systems: using electrophysiology in Xenopus laevis oocytes and using radioisotopes in S. cerevisiae.

 

 

Voltage trace showing depolarization of the membrane in response to the addition of D-glucose in the oocyte system.

Competition experiments using 300 µM D-glucose or 300 µM D-fructose and competitor in 10-fold excess in the yeast system.		Uncoupling experiments with 300 µM D-glucose as substrate in the yeast system.

These experiments provided the first conclusive proof that rust haustoria are indeed carbohydrate uptake devices.

Hexose Mobilization in Rust Fungi

Provision of substrate for the hexose transporter HXT1p is another question addressed in our lab. Glucose, one of the substrates of HXT1p, is not only an essential nutrient for most organisms, it also takes over signaling function. Consequently, the level of free glucose in any individual is tightly regulated. Now, how can the glucose transporter obtain its substrate? Mobilization of glucose may occur through the breakdown of sucrose (1-a-D-glucopyranosyl-2-b-D-fructofuranoside) the carbohydrate long distance transport form in most plants. We have good evidence that plant as well as fungal invertases are participating in this process. There are numerous reports of increased invertase activity upon wounding or pathogen infection in plants. However, no localization studies exist nor were there attempts to dissect the contributions of both organisms to the overall invertase activity. In the course of an EST sequencing project we have identified an Uromyces fabae invertase. At present we are working on a molecular and biochemical description of gene and gene product (Voegele et al. 2006). We are also studying the distribution of plant and fungal invertases within infected plants.

Hexose Utilization in Rust Fungi

The fate of carbohydrates once they are taken up by the fungus is also an important facet of our work elucidating the role of rust haustoria. From an EST sequencing project we have good evidence that glycolysis as well as pentose phosphate cycle are operating in haustoria. Currently we place the focus on the entry enzyme for these pathways, glucokinase / hexokinase [EC 2.7.1.2 / EC 2.7.1.1]. We want to find out which enzyme operates in haustoria, a hexokinase or a glucokinase. Substrate specificity of the entry enzyme is important since the hexose transporter transports fructose and glucose almost equally well. Are both sugars then funneled into metabolism with similar efficiency by the downstream enzyme, or is there a preference for one of the substrates? Preferential utilization of glucose through a downstream enzyme might lead to accumulation of fructose which would be an undesired physiological state. Or are there other ways to metabolize fructose?

A Link from Metabolism to the Suppression of Plant Defense in Rust Fungi

We have identified a mannitol dehydrogenase in U. fabae, MAD1/MAD1p. The enzyme inter-converts mannitol and fructose in an NADP(H)-dependent, reversible fashion. Fructose is one of the substrates of the hexose transporter HXT1p and mannitol is a C6-polyol with a number of interesting capacities. It can act a) as carbohydrate storage compound, b) as osmoprotectant, c) as radical scavenger, and d) in co-enzyme regulation and storage of reducing power (Voegele et al., 2005).

 

We were able to show an involvement of MAD1p and mannitol in carbohydrate storage and suppression of host defense reactions linked to reactive oxygen species.

Disappearance of mannitol from the fungal mycelium upon germination of the uredospores.		Quenching of ROS through increasing concentrations of mannitol.

Whereas MAD1 transcripts were only localized in haustoria, MAD1p was found in the lumen of haustoria and the lumen of spores.

MAD1 transcripts are only found in haustoria and infected leaves. A: schematic representation of rust infection structures. B: loading control. C: Northern Blot. The number on the right gives the size estimate in kb.		Localization of MAD1p in the lumen of haustoria and spores, using S717p. Superimposed phase contrast and fluorescence image. H: haustorium; S: spores (magnification 650-fold).

Our results classify MAD1p as an important multipurpose enzyme in U. fabae (Voegele et al., 2005).

Proteins from the Fungus Plant Interface

The interface between haustoria and the infected host cell represents a most ideal trading post not only for the exchange of nutrients, but also for the exchange of information. We therefore initiated a project to analyze the haustorial secretome, the entirety of proteins secreted from a haustorium. Our goal is to identify and characterize novel proteins, which might be linked to biotrophy or at least to the pathogenicity of rust fungi. Proteins for example like RTP1p, see below.

Rust Transferred proteins

The transfer of effector proteins from the pathogen to the host cell is a well accepted strategy for bacterial pathogens with hosts in the animal as well as the plant kingdom. However, the necessary 'type three secretion' or similar delivery systems have so far not been found in eukaryotes. It is nevertheless unlikely that fungal pathogens rely on the modulation of metabolite levels to influence host biochemistry and gene expression. In the course of screening the localization of proteins predicted to be secreted we have identified a protein, designated RTP1p, which is detectable beyond the outer limits of the extrahaustorial matrix (Kemen et al., 2005)

Superimposed phase contrast and fluorescence images depicting RTP1p localization
in the extrahaustorial matrix and in the cytoplasm and nucleus of the infected mesophyl cell.

Currently, we are trying to purify and characterize RTP1p, which does not exhibit sequence or structural homology to any protein deposited in the publicly accessible databases. Another aspect of our work on RTP1p is concerned with the mode of delivery for this protein to the host cell.

Rust Transformation

One essential drawback in our attempts to elucidate the role of haustoria is the lack of a functional stable transformation system for biotrophic fungi. Another aspect of our work therefore deals with establishing such a system. We have already established selection procedures on planta using the fungicides benomyl and carboxin (Wirsel at al., 2004). In addition we have started the construction of expression plasmids to be used in biobalistic or agroinfiltration experiments. We are confident to have this novel molecular tool for rust research available in the near future.

Methods

The expertise of our laboratory is the combination of a variety of molecular, biochemical and cytological tools, the latter in collaboration with Prof. K. Mendgen. Here is a short summary of methods used:

• PCR, mutagenesis and fusion PCR, semi-quantitative RT-PCR, real time PCR

• Southern and Northern hybridizations

• Protein purification by standard and FPLC techniques

• Heterologous expression of genes in different expression systems

	Escherichia coli	used for the production of antigens to obtain antibodies for cytological localization studies.
	Saccharomyces cerevisiae Schizosaccharomyces pombe	high level expression systems for cytoplasmic enzymes
	Pichia pastoris	high level expression system for the analysis of secreted proteins

• Antigen production for immunization

• Generation and purification of polyclonal antibodies

• SDS-PAGE and Western Blot analysis

• Light and fluorescence microscopy

• Electronmicroscopy