Plant development: Antibodies to the rescue

..::.;17"'-1_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ NEWSANDVIEWS ___________N_A_TIJ_RE_V_O_L..;...310_26--'-'-JU_L_Y_I984-'- Plant development Antibodies to the rescue from David Hanke IN biology, the most dramatic advances in understanding can generally be traced back to a new technique. Our knowledge of plant development looks to be poised fo( just such an expansion, judging by recent papers, including one on page 321 of this issue I, reporting results obtained using powerful immunological techniques. In the antibody we seem at last to have an experimental tool with the sensitivity and selectivity to match those of the control points in development. Many of these control points are sensitive to one or other of a clutch of simple organic chemicals, the plant growth substances, for which immunoassays have now been developed, most notably by Elmar Weiler2. Immunoassay provides a speedy, cheap and accessible technique, harnessing the astonishing selectivity of antibodies in order to detect and identify, and the affinity in order to measure, the traces of individual growth substances in plants. Antibodies have recently been used for the first immunocytochemical localization of a plant growth substance, dihydrozeatin, a cytokinin, in frozen sections of maize rooP. A particularly ingenious immunological method was used by Jacobs and Gilbert4 to locate the enigmatic biochemical machinery responsible for the fixed polarity of transport of the plant hormones named auxins. They raised a large number of monoclonal antibodies to total membranes from pea stems and selected out the cell clones producing antibodies which interfered most strongly with reversible binding to the membranes by naphthylphthalamic acid, a specific noncompetitive inhibitor of auxin polar transport. These monoclonal antibodies, specific for the naphthylphthalamic acid binding site, were then used to map the sites to the plasma membrane at the basal end of cells of the vascular parenchyma, exactly where the theories of polar transport had predicted they should be. More is known about the proteins that transport growth substances across membranes than about those that mediate their developmental effects. This is because whereas transport can be measured, the immediate functions of developmental receptors cannot, because we do not know what they are. We know that such receptors must exist because analogues of the growth substances are potent only if they mimic the genuine article with a steric precision, a singular disposition of chemical groups, that only a protein would be capable of appreciating. Now, Hornberg and Weiler I have achieved a major methodological advance, and quite literally got to grips with potential 'receptor' proteins, by brilliantly exploiting the unique chemical structure of another of the growth substances, abscisic acid. Until now the search for receptors has been carried out chiefly by looking for an appropriate concentration of 'sites' in the cell which bind the growth substance reversibly and saturably and with an affinity and specificity to match the characteristics of the response of the cell to the growth substance. The number and affinity of the binding sites is determined by analysis of the relationship between the amounts of growth substance bound reversibly at dif- o Structure of S-abscisic acid. The asymmetric carbon atom Col is its chiral centre. ferent concentrations s . This somewhat indirect method can generate bogus 'receptors'; for example, although only a single fusicoccin-binding protein can be isolated from plant cells, Scatchard plots of reversible fusicoccin binding to plant membranes are always biphasic 6 • Weiler and Hornberg have avoided this pitfall by covalently tagging the binding proteins with labelled growth substance. No cross-linking reagent was used, thereby greatly reducing the potential for artefacts, because abscisic acid itself can be induced by irradiation with UV light to irreversibly bind its 'receptor' through its ketone group at position C-4'. The method could hardly be more direct. The really clever part of the experiments, however, is the 'control'. The molecule of abscisic acid is chiral. In plants it occurs as the S-enantiomer only, while chemically synthesized material (much cheaper) is a racemic mixture which is difficult to resolve. Weiler and his colleagues have raised antisera specific for the R- and S-forms which enable them quickly and efficiently to purify the individual enantiomers. It is quite remarkable that immunoglobulins should be able to discriminate between the enantiomers because, as can be © 1984 Nature Publishing Group seen from the diagram, the molecule of abscisic acid is almost symmetrical on either side of a plane through the optical centre - the only difference between the two halves being tht C-6' carries two methyl groups whereas C-2' has one and a double bond 7 • Moreover, most plant tissues cannot distinguish between the enantiomers 8 • The best discrimination reported until now is by the stomata of barley leaves 9 • A stoma is an epidermal pore which is opened and closed by the expansion and contraction of two specialized cells which flank it, the guard cells. The responses of guard cells to growth substances are completely reversible and have to do with physiology rather than growth and development, but, rather like the thyroid and steroid hormones of animals, plant growth substances can affect many diverse processes. Thus an increase in the concentration of abscisic acid signals lack of water and both development and physiology may be adjusted accordingly. Working with broad bean leaves, Weiler and Hornberg find that R-abscisic acid has no effect on stomatal aperture over the first 30 minutes. This compound, which is virtually identical to the natural active form except in its chirality, is therefore a superb control. In the search for receptors only those guard-cell proteins that bind S-abscisic acid and not R -abscisic acid need be considered. In fact, Weiler and Hornberg find a large excess of cross-linking to proteins of guard-cell protoplasts by S- as opposed to R-abscisic acid. Similarities between the characteristics of the binding and the characteristics of the physiological response amount to an impressive correlation entirely consistent with the hypothesis that these binding proteins, anything up to three it seems, are involved in the response. Because proteolysed but intact protoplasts bind virtually no abscisic acid, Weiler and Hornberg argue that all the binding sites, which presumably include the receptors, are located on the outside of the plasma membrane. This is reasonable provided none of the proteins concerned is responsible for transporting S-abscisic acid across the plasma membrane ll and therefore also responsible for access to internal receptors. According to the only study so far of abscisic acid transport into guard cells, in this case of the unrelated wild flower Lamb's Lettuce, th (...truncated)