As plants are confined to the place where they grow, they have to develop a broad range of defence responses to cope with pathogenic infections. The oxidative burst, a rapid, transient, production of huge amounts of reactive oxygen species (ROS), is one of the earliest observable aspects of a plant's defence strategy. First this Review describes the chemistry of ROS (superoxide radical, hydrogen peroxide and hydroxyl radical). Secondly, the role of ROS in defence responses is demonstrated, and some important issues are considered, such as: (1) which of the ROS is a major building element of the oxidative burst; (2) the spatial and temporal regulation of the oxidative burst; and (3) differences in the plant's responses to biotic and abiotic elicitation. Thirdly, the relationships between the oxidative burst and other plant defence responses are indicated. These include: (1) an oxygen consumption, (2) the production of phytoalexins, (3) systemic acquired resistance, (4) immobilization of plant cell wall proteins, (5) changes in membrane permeability and ion fluxes and (6) a putative role in hypersensitive cell death. Wherever possible, the comparisons with models applicable to animal systems are presented. Finally, the question of the origin of ROS in the oxidative burst is considered, and two major hypotheses, (1) the action of NADPH oxidase system analogous to that of animal phagocytes, and (2) the pH-dependent generation of hydrogen peroxide by a cell wall peroxidase, are presented. On the basis of this material, a third ‘unifying’ hypothesis is presented, where transient changes in the pH of the cell wall compartment are indicated as a core phenomenon in evoking ROS production. Additionally, a germin/oxalate oxidase system which generates H2O2 in response to pathogenic infection is also described.

The origin of the oxidative burst during plant-pathogen interactions remains controversial. A number of possibilities have been identified, which involve the protoplast, plasmalemma or apoplast. The apoplastic production of H2O2 requires three components, an extracellular peroxidase, ion fluxes leading to extracellular alkalinisation and release of a substrate. Fatty acids are the major compounds that appear in the apoplast following elicitation, which can activate H2O2 production by peroxidases in vitro. However, the reaction with peroxidases appears to be novel and is uncharacterised at present. The apoplastic mechanism also cannot be readily distinguished from the operation of a plasma membrane NADPH oxidase system by the use of the inhibitors diphenylene iodonium and N,N diethyl-dithiocarbamate since it is also inhibited by these. These inhibitors have often in the past been used to define the involvement of the latter in the oxidative burst. In common with the NADPH oxidase system, the peroxidase responsible has been cloned but unlike the NADPH oxidase it has been shown to function in vitro to generate H2O2. In vivo studies of the oxidative burst have shown that the alkalinisation is essential and the underlying ion fluxes may be regulated by cAMP. Calcium fluxes are also essential. Although the oxidative activity of peroxidase requires calcium the fluxes have obvious other function. These may include activation of release of substrate and through the activation of a CDPK, regulation of enzymes involved in phytoalexin and cell wall phenolic production such as PAL.

NO is an important regulatory molecule in eukaryotes. Much of its effect is ascribed to the action of NO as a signalling molecule. However, NO can also directly modify proteins thus affecting their activities. Although the signalling functions of NO are relatively well recognized in plants, very little is known about its potential influence on the structural integrity of plant cells. In this study, the reorganization of the actin cytoskeleton, and the recycling of wall polysaccharides in plants via the endocytic pathway in the presence of NO or NO-modulating substances were analysed. The actin cytoskeleton and endocytosis in maize (Zea mays) root apices were visualized with fluorescence immunocytochemistry. The organization of the actin cytoskeleton is modulated via NO levels and the extent of such modulation is cell-type specific. In endodermis cells, actin cables change their orientation from longitudinal to oblique and cellular cross-wall domains become actin-depleted/depolymerized. The reaction is reversible and depends on the type of NO donor. Actin-dependent vesicle trafficking is also affected. This was demonstrated through the analysis of recycled wall material transported to newly-formed cell plates and BFA compartments. Therefore, it is concluded that, in plant cells, NO affects the functioning of the actin cytoskeleton and actin-dependent processes. Mechanisms for the reorganization of the actin cytoskeleton are cell-type specific, and such rearrangements might selectively impinge on the functioning of various cellular domains. Thus, the dynamic actin cytoskeleton could be considered as a downstream effector of NO signalling in cells of root apices.