Xyloglucans, xyloglucan endotransglucosylase (XET) activity, and arabinogalactan-protein (AGP)-like molecules: a new inter-relationship

The main resources generated were materials and methods for our own use and the use of others working on xyloglucan and the enzymes that act on it. We synthesised radioactively and fluorescently labelled oligosaccharides of xyloglucan. The methodology required for the fluorescent labelling was published in "Carbohydrate Research". We devised methods for the extraction of AGP-like carbohydrates, mainly from cauliflower florets, and methods for purification of AGPs from the crude extract. The work produced new, simplified assays for XET activity, and for the adsorption/desorption of XTHs to/from various surfaces, including glass, plastics, and natural and artificial polysaccharides. A method (involving the use of poly-lysine) for preventing the artefactual adsorption of XTHs to glass and plastic surfaces was developed. The fluorescently and radioactively labelled substrates produced during this project have been made available to other labs.
The simplified and highly sensitive assays developed in and around this project have been exploited by several collaborators. Several joint manuscripts involving the Edinburgh group (some/all of Fry, Sharples and Miller) plus a collaborating lab have been published bearing acknowledgement of BBSRC support. The lead labs on the various joint manuscripts are based in Cordoba, Argentina; Athens, Greece; Barcelona, Spain; Edinburgh, UK; and Southampton, UK.

Each cell in a plant has an outer wall, which may be either extensible (allowing cell growth, e.g. in a young beanstalk) or not (e.g. in an old leaf). We aimed to discover how the plant naturally changes its cell walls (making them more, or less, extensible) and thus control cell growth.

Walls of young plant cells (those potentially able to grow) are built of several kinds of polysaccharide chain, including cellulose and xyloglucan. Our lab is at the forefront of discovering, and understanding the role of, enzymes such as xyloglucan endotransglucosylases (XETs), which can cut and re-join xyloglucan chains in the walls of living plant cells. Such cutting/re-joining may be necessary (1) for new xyloglucan molecules to become built into the wall, and (2) for the wall to loosen its structure, enabling growth. We have also recently discovered, in horsetails, a novel enzyme that cuts a different polysaccharide (mixed-linkage glucan) and joins one end of it on to xyloglucan. This enzyme may strengthen ageing stems.

Other work had highlighted another class of polymers in and around the plant cell wall: 'arabinogalactan-proteins' (AGPs), complex polymers made of carbohydrate and protein. Various experiments had shown that AGPs somehow increase the ability of the wall to stretch and may thus somehow enable cell growth.

Our HYPOTHESIS on starting this project was that AGPs act by increasing the rate at which XETs cut and re-join xyloglucan chains. We had shown that some AGPs and related molecules can indeed promote XET activity in vitro. During the present project, we strengthened and extended the evidence for this hypothesis. We obtained AGPs and XETs from various plant species, and measured how fast each XET catalyses its cutting/re-joining reaction in the presence of each different AGP. We found that cauliflower AGP-like substances very strongly promoted XET activity, whereas acacia AGP did not. Interestingly, the cauliflower substances worked only when the XET was present at low concentration. This initially puzzling observation was investigated in detail and shown to be because XETs bind to solid surfaces (e.g. plant cell walls; even glass and plastic), thus inactivating the enzymes. The AGP-like substances were able to release the adsorbed XETs, thus re-activating the enzymes. In addition, we showed that in healthy plant cells, much of the XET is firmly bound to the cell walls, and that the AGP-like substances can release the enzymes from their entrapment within the wall, thus activating them.

The results are exciting in showing that enzyme action in the wall of a living plant cell is governed not only by the rate at which the cells synthesise and secrete the enzyme but also by an ability of the AGP-like substances to manoeuvre the enzyme within the cell wall, releasing it from a locked-up, inactive form. This discovery will guide future work and focus more attention on how plants control their cell-wall enzymes’ action, rather than simply their production and secretion.

We further tested the original hypothesis by seeing whether 'Yariv reagents' (chemicals that inactivate AGPs) stop XETs working in living plant tissues. We discovered that, although Yariv reagents inhibit plant cell expansion, they do so via their toxicity rather than by specifically altering XET action.

The knowledge during this project gained focuses attention on the need to characterise chemically the newly discovered factors ('AGP-like substances') that liberate XET from its inactive, wall-bound condition. Such work will open the way towards identifying genes that regulate the plant cell wall's properties, and thus in the future could give us novel ways of controlling genetically, biotechnologically or herbicidally the growth and development of crops.

1 We discovered that the xyloglucan endotransglucosylase action of XTHs on xyloglucan is controlled by the enzymes’ adsorption to cell walls. Soluble XTHs specifically and avidly bind to cell walls, cellulose and even glass, inactivating the enzyme. Native walls contain bound, ‘latent’ XTHs, suggesting an in-vivo significance of binding/inactivation for growth control.

2 Many tissues, e.g. cauliflower florets, were found to contain an ‘XTH-activating factor’ (XAF) that prevents XTHs binding the wall (and thus maintains their activity) and promotes the desorption of those already bound (and thus re-activates them). XAF does not modify enzymological parameters e.g. Km, but acts principally by controlling adsorption/ desorption.

3 XAF resembles arabinogalactan-proteins (AGPs); however, ‘classical’ AGP purification didn’t enrich in XAF. Detergents only weakly mimicked XAF’s effects. XAF did not bind xyloglucan, nor did it affect xyloglucan–cellulose hydrogen-bonding. The project opens the way towards characterising a novel ‘non-genetic’ mechanism of growth control.