9.2.1 Signal targets: perception and signal transduction

Printer-friendly version


Having outlined how hormonal signals are produced and transported, we turn to further equally important questions: how do cells detect the presence of hormones, and so perceive changes in hormone concentrations? And then, having measured the signal strength, how is this information converted into developmental and biochemical responses? For the first clue, we turn back to the signals themselves: we know that only certain molecular structures are biologically active and small changes in these molecules can render them virtually inactive. This happens with addition of hydroxyl groups to certain positions of a gibberellin molecule, or by comparing trans-zeatin with its cis isomer (Figure 9.1). This tells us that the mechanisms for detecting hormones must have great discriminatory powers. Hormone detection involves specific proteins known as receptors, proteins being the only class of biological macromolecule that can generate the required precision of molecular shapes. Within their three-dimensional structures, receptor proteins have regions which can bind the active hormone. These binding sites are similar to those in enzymes which bind a substrate, the familiar ‘lock and key’ analogy. The difference with receptors is that we think no chemical reaction occurs during perception; the hormone remains as hormone throughout. This is an important test in receptor assays which usually involve a radioactive hormone: if the hormone is converted to other compounds, probably the binding activity is simply to the active site of a hormone-metabolising enzyme.

(a) Plant hormone receptors

Plant hormone receptor research was neglected for many years but has attracted renewed interest with the advent of new assays and molecular biology tools since the late 1980s. Compared with detailed information on control of abundance and activities of mammalian hormone receptors, the picture in plants is sketchy. The best-studied systems have been auxin and ethylene receptors, and some of the genes that code for receptor proteins have been isolated. For example, the ETR gene from Arabidopsis was suspected to be an ethylene receptor, but this was only confirmed when the cloned gene was expressed in yeast in which the ETR protein was able to bind ethylene (Schaller and Bleecker 1995).There is also new evidence for gibberellin and ABA receptors, but little definitive information on cytokinins. Most of the receptors appear to be located on plant membranes, especially the plasma membrane, and this is also common in animal systems. Some elegant work by Hooley et al. (1991) strongly suggests that the gibberellin-binding site is located on the outer face of the membrane, so that it actually picks up hormone signals outside the protoplast. In one experiment, Hooley et al. synthesised gibberellin molecules anchored to large Sepharose beads which were incapable of entering the cell, but which still stimulated alpha-amylase production in protoplasts prepared from germinating cereal seed aleurone cells (Figure 9.4a). Alpha-amylase is an enzyme responsible for hydrolysing starch to sugars and hence supplying germinating seeds with carbohydrate for energy and growth. In another experiment, Hooley and co-workers generated antibodies that mimicked the shape of gibberellin molecules (known as anti-idiotypic antibodies). These would be recognised by the gibberellin receptor binding site and therefore competed with the gibberellin molecules and inter-fered with amylase production. The bound antibody molecules were also able to agglutinate protoplasts (Figure 9.4). Taken together, these lines of evidence indicate gibberellin perception occurring at the plasma membrane surface.


Figure 9.4 Gibberellin receptors in cereal aleurone cells are almost certainly located on the plasma membrane, facing outwards. Two lines of evidence support this view, both making use of protoplasts, that is, cells with their walls enzymatically removed. (a) Data showing that gibberellin molecules covalently anchored onto Sepharose (a polysaccharide gel) beads were still effective at inducing α-amylase synthesis. The Sepharose beads were much too large to enter the protoplast. The conclusion is that aleurone cells have receptors that can perceive gibberellin arriving from outside the cell. (b) Antibodies generated that mimic the shape of gibberellin molecules, known as anti-idiotypic antibodies, caused protoplasts to agglutinate (top). The explanation is that the antibodies are sticking to outward-facing gibberellin receptor binding sites on the plasma membrane. Because each antibody molecule has two binding domains, they can cross-link between cells, so forming aggregates of cells. Scale bar = 100um.

((a) Based on Hooley et al. 1991 and (b) based on Hooley et al. 1990)

(b) Selective signal transduction pathways

There is now the question of how a single hormone can be involved in several unrelated processes within the same plant. How do tissues stipulate the right response, at the right level and time? There are two main possibilities for ensuring that a signal is passed down the appropriate channel. Note that it is sometimes argued that plants may have only limited control over hormone production and supply and that the signals move and even act in a quite unpredictable manner through-out the plant.

Idea 1

For each response, there is a discrete type of receptor. Bearing in mind the many responses to each hormone, the total number of receptor forms would be high, but this does not necessarily rule out the idea. For example, there are at least five different genes for ACC synthase and three for ACC oxidase, both key enzymes in ethylene biosynthesis (Barry et al. 1996; Olson et al. 1991; Yip et al. 1992). Each member of the gene family is regulated by a different set of factors; thus some operate only in ripening fruit, others are induced by O2 deficit, others are switched on by auxin or by wounding. This division of labour at the hormone biosynthetic level may likewise occur in receptors, as shown by sequence homology between the ethylene-binding protein gene Etr1 and at least two other genes in Arabidopsis and five in tomato (Chao et al. 1997; Payton et al. 1996). Circumstantial support for multiple receptors also comes from the wide range of effective plant growth regulator concentrations, for example 10–10 M auxin promotes root elongation, compared with 10–6 M for the same process in shoots, and >10–4 M stimulates adventitious root initiation in stem cuttings. How could a single protein have the kinetic power to resolve concentration differences over more than a million-fold range? Auxin inhibition of growth may operate via ethylene because auxin at moderate to high concentrations induces ethylene synthesis (Sakai and Imaseki 1971). However, several other auxin responses are known to be ethylene in-dependent based on studies with auxin-overproducing Arabidopsis plants crossed with ethylene-insensitive mutants (Romano et al. 1993).

Idea 2

There are very few receptor types, possibly just one, for each hormone. The divergence of signalling would therefore occur ‘downstream’ from the receptor, that is, events that occur after hormone perception. There are several steps between reception and the final action, say, in closing a stomatal pore, inducing onset of dormancy or modifying a cell’s growth rate. In this way, a single ABA receptor could be coupled to different responses in guard cell, maturing seed and growing leaf, respectively.

Viewing the collective evidence, a tentative impression would be that perhaps both the above ideas are at least partly correct. Whatever the details of the system, in all cases the signal needs to be converted into a response. After perception by the receptor, the ‘signal’ is passed to what is commonly termed a second messenger, which in animals can be simple molecules such as Ca2+ ions, cyclic AMP (cAMP) or inositol trisphosphates (IP3). These are collectively involved in signal transduction pathways and usually involve some kind of amplification: from each hormone molecule binding to a receptor, many second messenger molecules may result. Typically, activation of a single enzyme molecule leads to generation of many product molecules, or opening of a single Ca2+ ion channel leads to flux of large numbers of Ca2+ ions. Animals and plants share some similarities in signal transduction mechanisms. Membrane-bound mammalian receptors are often linked to other proteins which, for example, bind GTP. These G-proteins are linked in turn to enzymes such as phospholipase C, which cleaves phospholipid groups and thus generates lots of second messenger product (IP3 and diacylglycerol) for as long as the receptor binding site is occupied. The number of G-proteins known in plants is increasing rapidly and they appear to have diverse roles in signalling (Millner and Causier 1996), including hormone systems such as ABA-regulated gene expression in germinating cereals (Bethke et al. 1997). In one of the best-studied hormone signalling systems, gibberellin induction of a-amylase gene expression in cereal aleurone cells, there are six or seven stages between primary signal and final action, namely production of active a-amylase enzyme (Figure 9.5; Bethke et al. 1997).


Figure 9.5 A complex chain of events is required to convert a primary hormonal signal, gibberellin, into its final action, in this case synthesis of the starch-hydrolysing enzyme α-amylase during cereal seed germination. After increased gibberellin levels are perceived, the most rapid changes, within 5 min, are in the second messenger Ca2+ ([Ca2+]i is cytosolic calcium concentration) then its receptor protein calmodulin (CaM), together with altered intracellular pH, and after about 1h another second messenger, cGMP. Following this, at abour 3h, there is a rise in GAMyb, a transcription factor protein that binds to promoters of gibberellin-regulated genes, and finally protein that binds to promoters of gibberellin-regulated genes, and finally appearance of α-amylase enzyme about 8h after gibberellin treatment

(Based on Bethke et al. 1997; reproduced with permission of Oxford University Press)

Within minutes, ionic changes (Ca2+ and pH) are detectable in the cytoplasm, followed by an increase in calmodulin, a calcium-binding protein involved in signal transduction. A slower increase in cGMP is seen which seems to operate independently of the Ca2+ system, but both of these transduction pathways seem to combine to stimulate transcription of genes via a protein called GAMyb, which is a transcription factor (a protein class that binds to specific DNA sequences in gene promoters; see Section 10.3.1). Overall, it takes 4–12 h before much functional a-amylase is detectable.

A lengthy debate on whether cAMP, a ubiquitous second messenger in animal systems, was important to plants was finally resolved by conclusive data showing presence not only of cAMP but also some of the proteins with which it interacts during auxin-induced cell division (Trewavas 1997; Ichikawa et al. 1997). The multiplicity of signalling components in plant cells and the number of potential links and interactions are beyond the scope of this discussion. What is remarkable is that primary signals are ultimately coupled to the ‘right’ response, whether direct physiological changes or altered gene expression. Much of this fidelity depends on restricted intra-cellular distribution of signalling components. For example, many of the protein kinases and protein phosphatases that activate and deactivate other regulatory proteins are probably located on membranes or tied to the cytoskeleton and therefore will only respond to signals within their immediate cellular neighbourhood (Trewavas and Malhó 1997). On top of that, cell differentiation will lead to quite different signalling components in each cell type. It may, however, be some time before enough is known about these subtle systems to be able to make use of them, say, in modifying crop physiology.