The plant cell surface and plant development

In normal development in multicellular plants, the directions of cell division and of cell enlargement play essential roles in the form assumed by a plant and its organs. Tissue development requires that cellular controls such as division and enlargement act locally. Some tissues (such as the vascular tissues of roots and shoots) display radial symmetry, while others (such as the component tissues of angiosperm ovules) are symmetrical only about a longitudinal axis. However, the structural basis of pattern control and of cell determination cannot be explained purely in terms of the direction of division and of the composition of cell walls (affecting enlargement). Patterning requires that certain cells or cell aggregates close to apical meristems become determined to particular developmental fates, and then differentiate accordingly. The position of these cells in the plant body must be controlled carefully, since all of organogenesis depends upon their differentiation. Somehow, cells acquire positional information early on.

Analysis of the control of plant cell fate is an intractible experimental problem. Genes have been identified that are involved with the control of complex developmental processes such as flowering (Coen et al., 1991), but these appear to affect floral organ identity. Other genes encode structural proteins and glycoproteins in plant cell walls that are developmentally regulated and hence candidates for intercellular interactions (Hong et al., 1989; Keller et al., 1989; Ye & Varner, 1991), but these seem to affect the mechanical properties of the cells rather than control their fate.

Introduction

The root-nodule bacteria Rhizobium, Bradyrhizobium and Azorhizobium (collectively rhizobia) invade and nodulate the roots of their host plants via either wounds or root hairs. The choice is made by the host plant, e.g. the same rhizobial strain infects Vigna roots via root hairs and Arachis roots via wounds (Sen & Weaver, 1984), whereas another strain infects Parasponia via root epidermal cracks and Macroptilium via root hairs (Marvel et al., 1985). Shortly before or during root invasion, rhizobia induce cell divisions in the root cortex, resulting in formation of a nodule primordium. Through infection threads (tip-growing tubular structures containing invading rhizobia) and/or between cortical cells the rhizobia migrate towards the growing primordium, are endocytosed by young nodule cells, and differentiate into dinitrogenfixing bacteroids (see also Brewin et al., this volume).

Rhizobial invasion of most agronomically important legumes such as pea (Pisum sativum), soybean (Glycine max) and bean (Phaseolus vulgaris) occurs through root hairs. Infection of a living plant cell is an unusual phenomenon in plant–bacteria interactions. Plants are open organisms. At many sites, the intercellular space of a plant is in direct contact with the environment, e.g. in stomata, hydathodes or wounds resulting from emergence of lateral roots. A plant is used to regular visits of (plant-associated) bacteria to its interior. Therefore, wound-infection by rhizobia is a normal phenomenon whereas root hair infection is special.

Introduction: Fucus as a model system

Our understanding of the molecular basis of recognition and signalling systems in plant cells is very poor, and information and concepts are often uncritically extrapolated from studies on animal cell systems. Most higher plant systems suffer from disadvantages such as limited accessibility to plasma membrane-based receptors because of intervening cell walls, thus necessitating protoplasting, with potentially undesirable side effects. Responses to stimuli are often slow, rendering experimentation more difficult. The analysis of the molecular organisation of plant plasma membranes has in itself been fraught with considerable difficulty because of problems in obtaining sufficient quantities of pure material and the lack of truly unequivocal criteria for purity.

In studies on higher plant reproductive systems, interactions between egg and sperm cells are particularly difficult to follow because the events are embedded within tissues, and although male and female gametes can now be isolated from angiosperms (for a recent review, see Theunis et al., 1991), this is still fraught with considerable technical difficulty. In this context then, ‘simpler’ recognition systems presented by oogamous lower plants such as Fucus have much to offer. Naked gametes of these algae are released in large enough quantities to permit detailed biochemical studies. The interaction provides a rapid and simple bioassay permitting direct analysis of the activity of blocking agents such as antibodies, oligosaccharides, putative receptor fractions, etc. Reproductive mechanisms apart, studies on Fucus gametes provide information on the molecular architecture of natural protoplast surfaces (e.g. the existence of discrete subsets (glycoforms) of glycoproteins, types of oligosaccharide structure and presence of topographical domains) that will be of value in exploring the organisation and complexity of plant plasma membranes in general.

Mycorrhizae are the most widespread type of association established between plants and soil microorganisms. The roots of 90% of land plants associate with many soil fungi to form complex systems, whose structure and function depend on the specific combination of the eukaryotic partners. During the formation of mycorrhizae, numerous events of specificity and recognition occur at different levels: species, plant organ, root tissue and cell type.

Specificity and recognition in mycorrhizae

Among the different 260 000 plant species, 200 000–240 000 have been estimated to have the potential to form mycorrhizal associations (Law & Lewis, 1983). On the other hand, many fungal representatives of the Zygomycotina, Ascomycotina, Basidiomycotina and Deuteromycotina depend on the mycorrhizal association for the completion of their life cycle (Harley, 1989). The single word ‘mycorrhiza’ encompasses a complexity of forms of interaction, since at least six main types of mycorrhiza can be recognised (Table 1). Extensive observations indicate that some specificity exists in the interaction, as a range of compatible and incompatible hosts can be defined. According to Smith & Douglas (1987) specificity of a symbiosis refers ‘to the degree of taxonomic difference between acceptable partners and may vary from very low to high or very high’. From Table 1 it is clear that mycorrhizae have a low degree of specificity compared to other associations, as a single fungus may associate with plants of more than one class and also a single host plant may associate with different fungal endophytes.

Introduction

There is a basic incompatibility between higher plants and fungi at the species level, an incompatibility which is the norm in interactions between these two types of organism. Plants in general are resistant to most fungal species, even those that are pathogenic on other plants. This type of incompatibility is synonymous with non-host resistance. It is also sometimes termed ‘heterologous incompatibility’ to distinguish it from the specific incompatibility observed in homologous interactions between plants and pathogenic fungal species (Gabriel & Rolfe, 1990). In this chapter I summarise what we know about the recognition and response events which appear to be involved in this basic rejection of fungi by higher plants. The study of ‘higher’ levels of specificity (e.g. the specific incompatibility between races of some biotrophic pathogens and certain cultivars of their hosts) has perhaps attracted more attention, but it can be argued that a better understanding of basic incompatibility would enable more rapid progress to be made in understanding the mechanisms underlying successful infections (basic compatibility) and from thence the higher levels of specific resistance involved in homologous incompatibility.

Defence mechanisms controlling incompatibility

What types of mechanism operate at the level of basic incompatibility to inhibit fungal growth? Although we now seem to know much about the nature of the individual mechanisms that can operate during these interactions, there is certainly much that we do not understand. Plants have a battery of potential defence mechanisms against non-pathogens that differ from species to species, although themes are evident.

Introduction

The fungi exhibit a polyphyletic range of sexual differentiation, with a variety of mating systems promoting genetic exchange and gene flow (see Carlile, 1987; Prillinger, 1987). This review considers examples of the biochemical and cytological mechanisms of fusions of cells of different mating types (Table 1).

Membrane fusion of motile gametes

Allomyces macrogynus, a member of the Chytridiomycetes, produces five types of flagellate cell in its life cycle: male and female gametes, the resultant zygote, and haploid and diploid zoospores. Of these, fusion occurs only between a male and female gamete. Both gametes are motile by means of one posterior flagellum, but there are many differences between them that must be phenotypic sexual differences, since they are both produced by the same haploid plant and must be isogenic. Only the female produces the chemotactic attractant, sirenin, and only the male (the spermatozoid) is attracted by it (Carlile & Machlis, 1965; Pommerville, 1978, 1981). Likewise, but less obviously, the male produces an attractant, parisin, which attracts the female (Pommerville & Olson, 1987). Sirenin is a sesquiterpene (Fig. 1d) and parisin shows similar chemical characteristics. The male gamete swims much more actively than the female, is bright orange with an accumulation of γ-carotene, is much smaller than the female, and has fewer and smaller mitochondria (Fig. 1a,b). It also appears to lack the ‘sidebody complex’ or ‘Stüben body’, a complex of lipid granules and microbodies of unknown function that is present in the female, and has fewer γ-like particles (Pommerville & Fuller, 1976).

Introduction

Nitrate from the soil is the main nitrogen source for most higher plants (Beevers & Hageman, 1969; Guerrero, Vega & Losada, 1981; Blevins, 1989). To increase both the growth and yield of those many crops that are unable to fix atmospheric nitrogen, farming has made use of nitrate-providing biological by-products or chemical fertilisers. There is a need to improve the control of the level of nitrate in the soil, to avoid water and atmospheric environmental pollution, as well as to lower production costs (Crawford & Campbell, 1990). For health concerns in humans and animals, there is also a need to maintain a low level of nitrate in food and forage plants. For these reasons, besides soil science, understanding which factors are involved in nitrate uptake and assimilation by plants and how these systems operate is of importance.

Nitrate absorbed by roots is assimilated inside the plant cell in the cytoplasm (see Fig. 1) either after transport from the outside through the plasma membrane, or through the tonoplast from the vacuole, where large amounts of nitrate can be accumulated. Both processes are active, and unfortunately remain poorly understood at a molecular level (Wray, 1988; Crawford & Campbell, 1990). Nitrate assimilation inside the cell involves two enzymes, nitrate reductase (NR), which reduces nitrate to nitrite, and nitrite reductase (NiR), which reduces nitrite to ammonium. This overall process is an 8-electron reduction step.

Introduction

The heat shock (hs) response in living cells is a fascinating subject for studying molecular mechanisms of stress-dependent regulation of gene expression and its phenotypic consequences (for reviews see Schöffl, 1988; Neumann et al., 1989; Morimoto et al., 1990; Nover, 1990). Particularly important is the transient acquisition of increased thermotolerance following a sublethal heat stress. Although exceptions occur, a positive correlation exists between the amount of hs proteins (hsps) synthesised in response to hyperthermia and the degree of tolerance. The synthesis of hsps seems to be necessary but may not be sufficient for thermotolerance. In different organisms different hsps are important for thermotolerance; the inability to synthesise hsps is usually correlated with thermosensitivity and an inability to acquire thermotolerance.

An example of a naturally occurring inability to synthesise hsp60, the sole hsp of the genus Hydra, is found in the thermosensitive species H. oligactis which is restricted to an ecological niche where it rarely encounters increasing water temperatures (Bosch et al., 1988). Mutations negatively affecting the hs response and thermotolerance properties have been described for Dictyostelium, Tetrahymena, Escherichia coli and yeast (for reviews see Neidhardt et al., 1984; Lindquist, 1986; Lindquist & Craig, 1988). A contrasting phenotype is exhibited by double mutations in two ubiquitin-conjugating enzymes of yeast which lead to the constitutive expression of hsps (Jentsch et al., 1990).

Introduction

Gene transcription requires the interplay between transcription factors with their cognate promoter elements. Structural and functional analyses of many of these transcription factors revealed a modular protein structure, composed of a DNA-binding domain and a transcriptional activation domain (Johnson & McKnight, 1989; Mitchell & Tjian, 1989). The DNA-binding domain of the b-Zip proteins is characterised by the presence of a basic region with an adjacent leucine zipper (Landschulz et al., 1988). Whereas the basic region is required for specific protein–DNA interaction and directly contacts the DNA, the leucine zipper facilitates homodimer and heterodimer formation (Hu et al., 1990 and references therein). Transcriptional activation domains are often enriched in acidic amino acids (Hope & Struhl, 1987; Ptashne, 1988), glutamines (Courey et al., 1989) or prolines (Mermod et al., 1989).

Although many DNA-binding proteins have been identified in plant nuclear extracts, only a few cDNA sequences encoding these proteins have been cloned. Plant b-Zip proteins for which cDNAs have been isolated include the wheat proteins, EmBP-1 (Guiltinan et al., 1990) and HBP-1 (Tabata et al., 1989), the maize proteins OCSBF-1 (Singh et al., 1990) and Opaque2 (Schmidt et al., 1990) and the tobacco proteins TGA1a and TGA1b (Katagiri et al., 1989). HBP-1, as originally identified in crude nuclear extract, was shown to interact with the hexamer (Hex) motif TGACGT found in several histone promoters (Mikami et al., 1987, 1989a,b,c). The cDNA identified as encoding HBP-1 was isolated by screening an expression library for specific DNA binding to an oligonucleotide derived from the wheat histone 3 promoter and containing the conserved Hex motif (Tabata et al., 1989).

Introduction

The phenylpropanoid pathway leads to the biosynthesis of a large number of phenolic compounds in plants. The importance of this diverse class of chemicals originating from phenylalanine has been recognised for some time, as it plays a key role in plant development and protection against environmental stress. The variety of chemical structures synthesised includes compounds like quinones involved in electron transport (French et al., 1976), flavonoid pigments responsible for flower coloration (Ebel & Hahlbrock, 1982) and protection against UV irradiation (Hahlbrock et al., 1982), and cinnamic acid esters and phytoalexins involved in disease resistance (Dixon et al., 1983). Phenolic metabolites like acetosyringone activate bacterial genes such as the Agrobacterium virulence genes or Rhizobium nodulation genes in the rhizosphere (Downie & Johnston, 1986; Stachel & Zambryski, 1986). Salicylic acid, which is involved in systemic induced resistance, is induced after wounding. Lastly, lignin, a major cell wall polymer found in close association with cellulose fibres and hemicellulose in the xylem, is a product of the phenylpropanoid pathway (Lewis & Yamamoto, 1990).

The biosynthesis of some of these compounds, for example those involved in flower pigment biosynthesis, is well understood. In addition, the structure and pattern of expression of genes encoding these enzymes has been studied extensively. The application of novel techniques which permit the modulation of gene expression has recently led to the manipulation of genes involved in flavonoid biosynthesis in Petunia.

Introduction

The alterations in colour, flavour, texture and aroma that are responsible for transforming an unripe, unpalatable fruit into one that is attractive to a consumer are known collectively as ripening. These ripening changes are catalysed by specific enzymes that cause biochemical modifications required for the ripening process to occur (Tucker & Grierson, 1987). In various fruits, different biochemical pathways have been recruited to the ripening programme during the course of evolution, accounting for the molecular, cellular, and anatomical differences between the many different species that produce fleshy fruits. Thus some fruits undergo no colour change (e.g. kiwi fruit), whereas in others the change in colour may occur in surface layers (e.g. some apples) or throughout the flesh (e.g. tomato). Different mechanisms operate for achieving the same ends, with generation of colour occurring either by the accumulation of carotenoids in chromoplasts (e.g. tomato, banana), anthocyanins in vacuoles (e.g. strawberry, blackcurrant), or the generation of special structures for thin-film interference of light, as in Elaeocarpus (Lee, 1991). During the onset of ripening, in addition to the generation of new structures and compounds, some existing ones such as chloroplast thylakoids and starch grains disappear. These changes are sometimes related to colour production. However, solubilisation of starch and cell walls contributes also to alterations in texture, taste, and juiciness. Further biochemical processes generate a balance of various organic acids and sugars and also produce aromatic compounds which confer a distinctive flavour and aroma to particular fruits.

Introduction

Crassulacean Acid Metabolism (CAM) has been called a curiosity (Osmond, 1978), the importance of which has been long overlooked in the context of environmental adaptations of plants to cope with arid, hot environments. CAM has since been studied more intensely and our knowledge has been extended by focusing on physiological studies (Ting & Gibbs, 1982; Ting, 1985; Cockburn, 1985), on the biochemical characterisation of important enzymes (O'Leary, 1982; Nimmo et al., 1986), and on ecological aspects (Kluge & Ting, 1978; Lüttge, 1987).

As a result of the diurnal separation of night CO2 fixation by PEPCase, storage of the acidic product, malate, in the vacuole, and final carbon assimilation by Rubisco, CAM plants conserve water and hence can occupy ecological niches that have limited water or CO2 availability. To mention only a few examples (see Kluge & Ting, 1978), CAM is expressed in flowering plants that are continually exposed to sea water, in cacti that inhabit true deserts, in climbing vines in the rainforest (Ting et al., 1985), and in freshwater plants where CO2 is limiting (Keeley & Busch, 1984). CAM may be constitutive in a species, or the pathway may appear during ageing either throughout the plant, or specifically in the older leaves (Guralnick et al., 1984; Ting, 1985). In some plants the pathway may be induced by environmental factors, such as drought, high salinity or low temperature, which affect the availability of water (Winter, 1982).

Plants are able to respond and adapt to changing environmental and endogenous signals by the induction of net synthesis of specific proteins which act to modify cellular metabolism. This text, based on papers presented at the Birmingham meeting of the Society for Experimental Biology in April 1991 in a two-day Symposium entitled ‘Biochemistry and Molecular Biology of Inducible Enzymes and Proteins in Higher Plants’, attempts to discuss the role these inducible proteins play in the biochemistry of the cell and the approaches being used to delineate the underlying molecular events which lead to their synthesis.

The topics included in this text do not exhaustively cover all the known responses of plants to environmental and endogenous signals, but a wide range are discussed. Whilst classical approaches, such as protein purification and characterisation, are of great importance in understanding the biochemistry of inducible proteins the molecular cloning of the encoding genes means that transgenic approaches can also be used to unravel the role particular proteins play in metabolism. In particular, the use of antisense RNA to down-regulate the synthesis of specific target enzymes is a particularly powerful technique and its use is discussed here in chapters on nitrate reduction, heat shock, phenylpropanoid biosynthesis and fruit ripening. The way in which transgenic techniques might be used in a broader sense to modify cellular metabolism to man's advantage is also discussed (see for example the chapter on anaerobiosis).

Introduction

Abscisic acid (ABA) mediates embryo maturation during late seed development. Maturation involves various morphogenic and biochemical changes including the programming of embryo dormancy and desiccation tolerance. Genetic analysis implicate ABA in the control of dormancy (Koorneef et al., 1984; McCarty et al., 1989). Molecular studies suggest that certain ABA-responsive genes expressed during late embryogenesis are part of a developmental programme leading to desiccation tolerance (Bartels et al., 1988; Dure et al., 1989). Certain of these genes are also expressed in vegetative tissues during osmotic stress (Mundy & Chua, 1988; Gomez et al., 1988), at which time ABA levels rise and growth is inhibited. At present, the connection between ABA effects during embryogenesis and in vegetative tissues is unclear, in part because we do not know the function(s) of most of the major ABA-responsive genes (for reviews, see Skriver & Mundy, 1990; Galau et al., 1991; McCarty & Carson, 1991). However, the hormone mediates responses to osmotic stress, and causes developmental or growth inhibition in both embryonic and vegetative tissues (Smart & Trewavas, 1984; Bensen et al., 1988; Creelman, 1989).

In many fruits, embryo dormancy is broken by environmental cues and physiological factors which initiate germination. In cereal seeds, gibberellic acid (GA) appears to be one such factor by promoting this expression of genes encoding α-amylase and other hydrolases. This effect of GA can be inhibited by ABA (Jacobsen & Beach, 1985; Nolan & Ho, 1988; Huttly & Baulcombe, 1989), an antagonism which may mediate physiological changes controlling the switch from seed quiescence or dormancy to germination.