1Laboratorium voor Genetica, Departement Genetica, Viaams lnteruniversitair lnstituut voor Biotechnologie (VIB), Universiteit Gent, KL Ledeganckstraat 35, B-9000 Gent, Belgium; 2Vakgroep Plantaardige Productie, Faculteit Landbouwkundige en Toegepaste Biologische Wetenschappen, Coupure Links 653, B-9000 Gent, Belgium

Background and objectives
Sedentary endoparasitic nematodes have evolved a parasitic lifestyle that is based on an intimate relationship with the host plant. Two different types of endoparasitic nematodes can be distinguished: root-knot and cyst nematodes, both inducing a feeding site in the plant root, but in a different way. To obtain insight into the molecular mechanisms behind this complex interaction, several strategies to analyse plant gene expression in response to nematode infection have been followed. Concomitantly, the signals coming from the nematode that are triggering this plant response are being characterized.

Results and conclusions
The mechanism of nematode feeding site induction is unknown, although components of nematode secretions most probably act as triggers. Cyst nematode secretions induce a process of cell wall breakdown between the initial cell of the feeding site and neighbouring cells in the developing vascular cylinder. The result is a syncytium with multiple large nuclei due to endo-reduplication. The response of plant cells to root-knot nematode secretions is different. One of the first detectable changes is the repeated division of the nucleus without cell division. This leads to a large multinucleated cell called a giant cell. The feeding site of a root-knot nematode is composed of several giant cells surrounded by a gall. One of the earliest responses of the plant root cells to the nematodes that is common to both types is at the nuclear level. Both types of feeding sites contain multiple enlarged nuclei, and the underlying mitosis and/or endo-reduplication is supported by the expression of several cell cycle genes [1]. The establishment of the feeding sites has to be governed by differential regulation of many other plant genes. Approaches to identify these genes involve comparisons between infected and control roots, including protein analysis, construction and screening of cDNA libraries, differential display of transcribed sequences and the analysis of reporter genes in transgenic plants, either fused to known promoters or in a promoter-trap strategy. Random in vivo gus fusions have been particularly successful in identifying plant promoter sequences that are highly activated in nematode feeding sites, with very little expression elsewhere in the plant, but the isolation of the corresponding genes is usually not that straightforward [2]. Differential display enables the direct isolation of differentially expressed genes; however, the cellular expression pattern needs then to be analysed by in situ hybridizations. Many different plant genes have already been identified as highly transcribed in the feeding sites, but few have been characterized in sufficient detail to know how important they are for a successful infection [3]. Intriguingly, several tested plant genes are not up-, but down-regulated in the nematode feeding sites. How this relates to inhibition of the plant defence response is an open question. In conclusion, we can say that a thorough analysis of the available plant genes that have been shown to be nematode-responsive will bring exciting new insights into the pathogenesis process.

1. Gheysen G, de Almeida Engler J, Van Montagu M, 1997. In Fenoll C, Grundier FWM, Ohi SA, eds, Cellular and Molecular Aspects of Plant-Nematode Interactions (Developments in Plant Pathology, Vol. 10). Kluwer, Dordrecht, pp. 120-132.
2. Barthels N, van der Lee FM, Klap J et al., 1997. Plant Cell 9, 2119-2134.
3. Fenoll C, Aristizobal FA, Sanz-Aifdrez S, del Campo FF, 1997. In Fenoll C, Grundier FWM, Ohi SA, eds, Cellular and Molecular Aspects of Plant-Nematode Interactions (Developments in Plant Pathology, Vol. 10). Kluwer, Dordrecht, pp. 133-149.