The goal of Dr. Kang's research program is to study plant- pathogen interactions at multiple levels, ranging from genes to ecosystems.

(A) Root pathogenesis/defense using Arabidopsis thaliana as a host: Because much of our current knowledge on the molecular basis of plant-fungal pathogen interactions has been derived from studies based on foliar pathogens, our understanding of soil-borne diseases lags considerably behind. My approach is to employ a diverse array of pathogens representing different infection strategies and tissue specificity so as to identify both general and pathogen-specific defense mechanisms in A. thaliana. Two soil-borne pathogens, Fusarium oxysporum and Verticillium dahliae, have been utilized in these studies. Members of the F. oxysporum species complex are the most common fungi in soil and consist of numerous pathogenic and non-pathogenic forms. Pathogenic members can be divided into many host specific forms (i.e., forma specialis = f. sp.) and collectively cause vascular wilt, damping-off, and root rot diseases in over 100 cultivated plant species. Rich strain resources (>1,000 isolates) in combination with well-established phylogenetic relationships within the species complex, nicely support studies on the evolutionary mechanisms underpinning host specialization. Verticillium dahliae also has a broad host range and causes vascular wilt, but unlike F. oxysporum , many isolates do not exhibit clear host specificity. The broad host ranges of F. oxysporum and V. dahliae make it possible to compare pathogenicity and defense mechanisms in diverse plant species.

We identified eight F. oxysporum isolates pathogenic to A. thaliana. Among 48 ecotypes infected with two of the isolates, substantial variation in resistance both within and between isolates was observed. Mechanisms of their interactions have been characterized using a combination of cytological, genetic, and metabolomics tools. Using two A. thaliana ecotypes, Cvi-0 (susceptible) and Gre-0 (resistant), three-dimensional, time-resolved data from individual infection sites by fluorescently-labeled fungal strains were obtained over several days without physical manipulation of infected plants (in collaboration with Dr. Kirk Czymmek at the University of Delaware). This technique allowed monitoring of fungal growth on the root surface and within the vascular tissue and observation of changes in root cells in response to fungal growth. Initial penetration occurred primarily in the meristematic region of primary and lateral roots and seems to require a mycelial mass for penetration (similar to quorum sensing in bacterial pathogens). The fungus appears to produce phytotoxin(s) that could effectively move to nearby cells and affect vacuolar membrane integrity and potentially induce host cell death.

(B) Rice blast: Worldwide, this disease, caused by Magnaporthe oryzae, is the biggest threat to rice cultivation. Besides its economic significance, rice blast presents many advantages as an experimental model. Unlike most plant.fungal pathogen systems, the genome sequences of both M. oryzae and rice are available, providing a unique opportunity to study their interactions from both sides using functional genomics tools. Two areas of my research on rice blast include the genetic mechanisms underpinning the breakdown of resistance in the field and the molecular and cellular basis of pathogenesis/defense. Although resistance ( R ) gene-mediated resistance is highly effective once triggered, such resistance frequently loses its effectiveness in the field due to the emergence of new races. Considering the importance of utilizing R genes to control various diseases, especially in many developing countries where chemical control is economically impractical, answering the questions of how new races arise and how gene-for-gene interactions are controlled is crucial for sustainable disease management. I have been pursuing these questions at multiple levels, ranging from genes to field populations. Specifically, I have focused on the evolution, function and variation of the PWL and AVR-Pita avirulence (AVR ) gene families in M. oryzae. Certain members of these gene families are specifically expressed in planta and seem to be transported to host cytoplasm via a vesicle-like structure (termed Effector Secretion Bodies) on infection hyphae (Fig. 1). Fungal genes controlling the expression and function of these gene families are currently being isolated.

Fig. 1 . Localization of the AVR-Pita1-GFP fusion protein in infection hyphae. (A) Bright field image of infection hyphae inside rice epidermal cells at 24h post inoculation. (B) Hypha in (A) visualized with the GFP filter. An effector secretion body (ESB) is denoted by an arrow. This work has been conducted in collaboration with Dr. Barbara Valent at KSU.

By taking advantage of genome sequences from M. oryzae , Dr. Yong-Hwan Lee at SNU and I have conducted a genome-wide search for pathogenicity genes, resulting in the identification of ~200 such genes. Functional characterization of these genes is in progress. Interactions between Arabidopsis thaliana and M. oryzae are currently being studied to dissect nonhost resistance. The fungus infects certain ecotypes, but the required factors for A. thaliana infection are different from those for rice infection.

(C) Cyber-infrastructure for Plant Pathogens (CiPP): The goals of CiPP are to integrate existing genotypic and phenotypic information on plant pathogens with important environmental variables, and to engage the global community of plant pathologists to use state-of-the-art data mining and visualization tools for the advancement of science, education, and outreach (Fig. 2). Our first and foremost reason for building CiPP is to collect and catalog biological information and materials for present and future use. Properly archiving accumulated data and materials in a format that can maximally support future research is as important as generating new data and materials. Because science builds on existing knowledge, failure to establish proper links between what has been done and what will be done is a poor scientific practice and frequently forces us to .reinvent the wheel.' Pathogen cultures often form the primary link that connects discoveries of the present with established knowledge of the past, facilitate comparisons of findings in different areas, and support pathogen forensics. The second motivation came from the realization that the full potential of pathogen genomics as a foundation for understanding and managing disease dynamics, hinges on how effectively we use genome sequence data to gain a comprehensive understanding of the evolutionary and pathological potential and mechanisms within species.

Fig. 2 . Organization and functionality of CiPP. The CiPP weaves together (i) Phytophthora Database; (ii) detailed global distribution maps of Phytophthora ; (iii) versatile molecular diagnostic tools; (iv) IT tools supporting the use of archived data for Phytophthora detection and identification; (v) GIS tools for visualizing the distribution and change of Phytophthora in environmental and geospatial contexts; and (vi) a globally-linked network of scientists working together in monitoring Phytophthora diseases.

The Phytophthora Database ( www.phytophthoradb.org ) is the first fruit of CiPP. Database for other groups of pathogens are in progress. Given increasing movements of pathogens via global agricultural trade, a pathogen monitoring system focused primarily on the United States would not be adequate. In collaboration with scientists in many parts of the world, we are building a global atlas of Phytophthora. In the long run, the database will support integration and utilization of data from diverse areas of research on Phytophthora, ranging from genomics, phylogenetics and population biology to epidemiology.

(D) Development of research tools and resources: During the last 3-4 years, we have developed experimental tools and resources that will help us study dynamic changes underlying plant-pathogen interactions at multiple levels, ranging from genes to whole organisms. One resource is the genome sequences of several fungal pathogens that, in combination with the genome sequences of host plants, will facilitate studies on their interactions from both the plant and pathogen sides. I am a coPI of two genome sequencing projects (funded by the USDA Microbial Genome Sequencing program), one for F. oxysporum and F. verticilliodes and the other for V. dahliae and V. albo-atrum. The former project is completed, and the latter project is in progress. Another area of investment is the development of cytological tools for both plants and pathogens to study cell-to-cell communications within and between plant and pathogen throughout the disease cycle. These tools include FRET-based biosensors for Ca and cAMP, pH-sensitive GFP and an array of fluorescent markers for labeling fungi and plants. Transformants of A. thaliana , F. oxysporum , and M. oryzae with some of these sensors have been generated, which are currently being used to map cell-to-cell communications during various stages of pathogenesis/defense. A third resource is fungal gene manipulation tools via the use of Agrobacterium tumefaciens.