Use of mutants to study long‐distance signalling in response to compacted soil

Abstract

When plants encounter compacted soil, stomatal closure occurs and shoot growth slows. These responses occur in the absence of detectable changes in foliar water status. The use of genotypes with a reduced capacity to synthesize either ABA or ethylene has provided convincing evidence that ABA is responsible for providing the signal that regulates stomatal aperture, whereas increased ethylene production leads to an inhibition of shoot growth. Compaction results in an elevated export of ABA from the roots while enhanced ethylene synthesis is associated with increased expression of ACC oxidase in the aerial parts of the plant. Future work will explore the mechanisms responsible for regulating these events and the contribution of anaerobiosis to the stresses experienced by roots growing under compacted conditions.

Introduction

When the roots of plants encounter soil of high bulk density, shoot growth slows and stomatal closure takes place (Masle, 1992; Mulholland et al., 1996a). Both phenomena occur prior to detectable changes in foliar water status (Tardieu et al., 1992; Andrade et al., 1993). These observations provide strong circumstantial evidence that one or more root‐sourced signals is responsible for eliciting the responses of the shoot to compacted soil and several possible candidates have been proposed (Sarquis et al., 1991; Hartung et al., 1994; Beemster and Masle, 1996; Masle, 1998). One of these, abscisic acid (ABA), is widely accepted to act in this capacity when roots experience drying soils (Davies and Zhang, 1991; Davies et al., 1994), while another, ethylene, is believed to play a central signalling role during the response of plants to waterlogging (English et al., 1995). In this review, the contributions of these two plant growth regulators (PGRs) in bringing about the syndrome associated with soil compaction will be evaluated. In addition, there is an examination of how the study of mutants has assisted the understanding of the events that take place. Genotypes with an attenuated capacity to produce PGRs provide a powerful tool for probing the role of these compounds in plant development and the availability of material deficient in either ABA or ethylene biosynthesis has been central to the success of these studies.

Stomatal responses

Role of ABA

When all or part of the root system of wild‐type Ailsa Craig tomato plants encounter compacted soil, significant reductions in stomatal conductance are observed. However, this response is not exhibited by the ABA‐deficient mutants notabilis and flacca (Mulholland et al., 1999; Hussain et al., 2000). As no effects on leaf water status were detected in any of these genotypes, these data support the hypothesis that ABA has a key role in controlling stomatal behaviour under compacted soil conditions. Further evidence supporting this assertion comes from studies of the wild‐type cultivar of barley, Steptoe, and its isogenic ABA‐deficient line Az34 (Mulholland et al., 1996a).

In order to examine the nature of the signalling events further, a split‐pot system was developed to investigate the importance of root‐sourced signals in co‐ordinating plant responses to soil compaction. In this system, tomato seedlings were initially established in an uncompacted soil horizon (1.1 g cm⁻³), but as the root system develops it is grown into two compartments (Hussain et al., 1999a; Mulholland et al., 1999); one contained uncompacted soil, whilst the other contained severely compacted soil (1.5 g cm⁻³). An additional ‘split‐cut’ treatment was applied in which the roots growing in the compacted compartment were severed at a specific time after seedling emergence. Use of this novel experimental strategy revealed that severing the roots growing in the severely compacted environment increased stomatal conductance in wild‐type plants relative to equivalent plants in which the roots growing in uncompacted soil were not severed, whereas no similar response was observed in ABA‐deficient mutant plants. These data provide additional support for the view that a root‐sourced signal is involved in mediating the observed reductions in stomatal conductance when roots encounter compaction and suggest that excision of the roots growing in compacted soil removes the flux of the signal to the shoot.

Analysis of xylem sap demonstrated that a 2–3‐fold increase in ABA concentration occurred when the roots of wild‐type plants encounter compacted soil, and a significant inverse correlation was obtained between stomatal conductance and xylem sap ABA concentration (Hussain et al., 2000). Taken together, these observations provide compelling evidence that a root‐sourced ABA signal regulates stomatal behaviour, a view further substantiated by the discovery that the split‐cut treatment resulted in a reduction in xylem sap ABA concentration that was closely correlated with the recovery of stomatal conductance.

Role of ethylene

Transgenic tomato plants (ACO1_AS), in which the conversion of 1‐aminocyclopropane carboxylic acid (ACC) to ethylene has been blocked (Hamilton et al., 1990), exhibited a similar pattern of stomatal behaviour to wild‐type plants grown under uniformly compacted soil conditions (Hussain et al., 1999b). Significant reductions in stomatal conductance were also observed in the split‐pot treatment of ACO1_AS, whilst excision of the roots in the compacted compartment led to a recovery of stomatal conductance. Increases in xylem sap ABA concentration were detected in the split‐pot and compacted treatments of ACO1_AS, whereas the values declined following excision of the roots growing in high bulk density soil. These observations for ACO1_AS plants indicate that endogenous ethylene levels have little influence on the ABA‐based regulation of stomatal behaviour in response to soil compaction. Further evidence to support this assertion comes from analyses of stomatal conductance in plants in which ethylene production was elevated by applying exogenous ACC or the ethylene‐releasing compound, ethephon. Blocking the action of ethylene with silver ions (Beyer, 1976), supplied as silver thiosulphate (STS), promoted patterns of stomatal behaviour and xylem sap ABA levels comparable to control plants supplied with water (Hussain et al., 1999b).

Growth responses

Role of ABA

When ABA‐deficient mutants of barley or tomato (Fig. 1A, C) were grown with all or part of their root systems in compacted soil they exhibited similar or greater reductions in leaf expansion than wild‐type plants (Mulholland et al., 1996a, 1999). These observations suggest that it is unlikely that the observed growth reductions are mediated by ABA. Indeed, in barley, there is convincing evidence that ABA has a positive role in maintaining leaf expansion at a ‘sub‐critical’ bulk density (Mulholland et al., 1996b).

Role of ethylene

As ethylene evolution from the leaves of wild‐type and ABA‐deficient mutant plants increases when all or part of the root system encounters compacted soil (Fig. 1D, F), it is possible that this gaseous PGR may contribute to the observed reduction in shoot growth. To examine the role of ethylene in mediating reductions in leaf growth, ACO1_AS plants, isogenic wild‐type and ABA‐deficient mutant (notabilis) plants were grown using the split‐pot system described above (Hussain et al., 1999a, 2000). Wild‐type and ACO1_AS plants exhibited comparable leaf growth rates when grown in uncompacted soil, whereas growth was consistently lower in notabilis. This is a common feature of ABA‐deficient tomato genotypes (Fig. 1A–C). All genotypes displayed significant reductions in leaf area when grown in uniform 1.5 g cm⁻³ soil. These data suggest that this level of compaction is too severe for tomato to tolerate and that reductions in ethylene or ABA levels have little impact on shoot expansion due to the extreme limitation on root growth. The split‐pot system induced phenotypic differences in response between the genotypes studied, with leaf growth being intermediate between the uniformly uncompacted and compacted treatments in wild‐type plants, but comparable to plants in the uncompacted treatment in ACO1_AS (Hussain et al., 1999a). Leaf growth in notabilis was greatly reduced in the split‐pot treatment, with values being closer to plants in the uniformly compacted treatment. The split‐cut treatment resulted in a recovery of leaf growth in both wild‐type and notabilis plants, but had little effect on ACO1_AS. These results therefore demonstrate the involvement of ethylene in mediating reductions in leaf growth under split‐pot conditions, since leaf growth was not suppressed in ACO1_AS but was reduced in wild‐type plants. Thus, impairment of the capacity of plants to synthesize ethylene resulted in an inability to regulate leaf growth under ‘sub‐critical’ compaction stress.

The hypothesis that the observed reductions in leaf growth in the split‐pot system were mediated by ethylene was supported by measurements of ethylene evolution from leaves. As anticipated, ethylene evolution was lowest in all treatments of ACO1_AS, whereas 3–4‐fold increases were apparent in the uniform 1.5 g cm⁻³ treatment of wild‐type and notabilis plants; these increases were accompanied by reductions in leaf growth (Fig. 1). The split‐pot treatment of wild‐type plants produced rates of ethylene evolution intermediate between those obtained for plants grown in uniformly uncompacted and compacted soil, reflecting the intermediate leaf growth rates; however, the same treatment of notabilis resulted in ethylene evolution rates comparable to the uniform 1.5 g cm⁻³ treatment. The split‐cut treatment invoked little response in ACO1_AS, but increased leaf growth in wild‐type and notabilis plants. Reductions in leaf growth were therefore closely correlated with ethylene production, particularly in the split‐pot treatment. Further analysis of these data revealed a significant inverse correlation between ethylene evolution and leaf area (Hussain et al., 2000).

Treatment with STS promoted a recovery of shoot growth when wild‐type and notabilis plants were grown in the split‐pot system, but had no effect on ACO1_AS. As the application of silver ions is known to block the action of ethylene, these results suggest that the subsequent recovery of leaf growth in wild‐type and notabilis plants resulted from a reduction in the inhibitory influence of ethylene. However, this treatment had little impact on ACO1_AS due to its limited capacity to produce ethylene (Hussain et al., 1999b). Treatment with ethephon increased ethylene evolution in all genotypes, but promoted much larger reductions in the shoot growth of wild‐type and especially ACO1_AS plants than in notabilis (Hussain et al., 1999b).

The role of ethylene in mediating leaf growth has been further investigated using transgenic plants impaired in the production of ACC synthase; this enzyme is responsible for forming ACC, the precursor of ethylene. These ACS_AS plants were compared with their isogenic wild‐type equivalent (VF36). When VF36 plants were grown in the split‐pot system described previously, leaf growth was comparable to Ailsa Craig, whereas the reduced ethylene synthesis associated with the decreased ACC production in ACS_AS resulted in leaf growth being comparable to ACO1_AS. Application of ACC to the compartment containing compacted soil produced a small increase in ethylene production in VF36 that was reflected by a small reduction in leaf growth. However, in the same treatment of ACS_AS, conversion of exogenous ACC to ethylene significantly increased ethylene production and greatly reduced leaf growth, further implicating ethylene in the control of leaf growth in compacted soil.

Regulation of ethylene biosynthesis

The involvement of ethylene in mediating leaf growth when roots encounter compacted soil has been further investigated by analysing the expression of the ACC oxidase gene family. In tomato there are four known ACC oxidase genes, ACO1, ACO2, ACO3, and ACO4 (Nakatsuka et al., 1998). Northern analysis of the expression of these genes in leaf tissue from Ailsa Craig, notabilis and ACO1_AS plants grown in the split‐pot system showed that the expression of one of these genes, ACO1, was up‐regulated when all or part of the root system encountered compacted soil in both wild‐type and notabilis plants. As anticipated, ACO1_AS plants exhibited a low level of expression of the ACO1 gene. The other three ACC oxidase genes showed no up‐regulation in response to soil compaction in any of the genotypes examined. These findings provide evidence that the observed increases in ethylene evolution may, in part, result from elevated conversion of ACC to ethylene; perhaps in an analogous way to that observed during waterlogging (English et al., 1995).

The formation of ACC is regulated by the ACC synthase gene family, which in tomato consists of seven known genes, ACS1, ACS2, ACS3, ACS4, ACS5, ACS6, and ACS7. Measurement of the expression of these genes in leaf tissue revealed that they were constitutively expressed and demonstrated no up‐regulation in response to compaction in wild‐type, notabilis or ACO1_AS plants. These observations suggest that the elevated ethylene production associated with plants subjected to soil compaction is a consequence of enhanced conversion rather than elevated synthesis of ACC in the leaf tissues, but do not preclude the possibility that the supply of ACC from the roots is increased; this question is currently under investigation. Results for plants from the split‐cut treatment suggest that leaf growth may be regulated by a root‐sourced signal capable of controlling ethylene production.

Interaction between ABA and ethylene

As increased xylem sap ABA concentrations are able to maintain leaf growth during ‘sub‐critical’ compaction stress in barley (Hussain et al., 1999c) it is possible that interactions between ABA and ethylene may be involved in mediating leaf growth when plants encounter compacted soil. To test this hypothesis, the interactions between ABA and ethylene were investigated for Ailsa Craig, ACO1_AS and notabilis. When plants were grown in the split‐pot system, leaf area in ACO1_AS was unaffected relative to the uncompacted treatment, but was intermediate between the uncompacted and compacted treatments in wild‐type plants, suggesting that this reduction in leaf area was mediated by ethylene. In notabilis, leaf growth was greatly reduced in the split‐pot treatment and leaf area was comparable to the uniformly compacted treatment. These results suggest that the more vigorous leaf growth in wild‐type plants relative to notabilis occur because the latter genotype is unable to synthesize wild‐type levels of ABA. Indeed, measurement of ethylene evolution from the leaves showed that the values for the split‐pot treatment of wild‐type plants were intermediate between those for the uncompacted and compacted treatments of the same genotype; by contrast, the values for the split‐pot treatment of notabilis were comparable to those in the uniformly compacted treatment. This observation suggests that the more rapid ethylene evolution in notabilis relative to Ailsa Craig is not only correlated with the greater inhibition of leaf growth in this treatment, but might also be a consequence of the lower endogenous ABA levels. Thus, the presence of wild‐type ABA levels during ‘sub‐critical’ soil compaction may limit ethylene production, thereby permitting more rapid leaf growth than in ABA‐deficient plants. Moreover, analysis of the data obtained revealed an inverse correlation between ethylene evolution and xylem sap ABA concentration for plants grown in the split‐pot treatment (Hussain et al., 2000).

Application of ABA to tomato plants grown in the split‐pot system had no effect on leaf growth in ACO1_AS, but invoked a marked increase in notabilis, for which the values were intermediate between the uncompacted and compacted treatments. The recovery of leaf growth following restoration of wild‐type xylem sap ABA levels by applying exogenous ABA was accompanied by a significant decline in ethylene evolution from leaf tissue to a level intermediate between the uncompacted and uniformly compacted treatments. This observation therefore provides evidence that wild‐type levels of ABA reduce ethylene production when plants encounter ‘sub‐critical’ soil compaction. Interestingly, exogenous ABA also promoted a small increase in leaf growth in wild‐type plants grown in the split‐pot system, reflecting a small reduction in ethylene evolution relative to ABA‐untreated plants; this finding further substantiates the hypothesis that ABA has the capacity to influence ethylene synthesis. The influence of ABA levels on ethylene production and action is of great interest and will be a focus for future investigations.

Conclusions

These studies have revealed that ABA and ethylene may both act as signalling molecules when roots experience compacted soil. It seems likely that elevated ABA is a long‐distance signal transported from the roots to the shoots in the transpiration stream and that more local increases in ethylene occur following enhanced conversion of ACC in the leaves. An important question yet to be resolved is whether roots growing in compacted soil export increased quantities of ACC to the shoots. If this proves to be the case, cells within the roots may therefore be induced to increase their synthesis of both ABA and ACC under compacted soil conditions. A target for future work will be to determine the spatial and temporal changes in gene expression responsible for these effects. Due to the severe practical difficulties involved in carrying out physiological and molecular studies of roots grown in soil‐based compaction systems, an alternative system comprising ballotini of differing size has been successfully developed to provide a range of resistances to root growth for plants continuously supplied with nutrient solution. As the growth medium may easily be removed, this approach provides easy access to the roots, thereby facilitating hormonal and molecular analysis.

In addition, as plants growing on compacted soils frequently experience the coupled effects of impeded root growth and anaerobic rooting conditions, this approach will make it possible to distinguish between these two stimuli by varying the oxygen content of the nutrient solution. Furthermore, in split‐pot experiments, it will be possible to supply the individual compartments with differing nutrient solutions, thereby permitting elegant feeding experiments involving the application of ABA or chemicals that influence ethylene action. It is anticipated, that by using these experimental procedures and by continuing to exploit the use of mutant and transgenic plants, it will be possible not only to identify the long‐distance signalling network induced by compacted soils, but also to ascertain how it is initiated.

Financial support from the UK Natural Environment Research Council and the Biotechnology and Biological Science Research Council is gratefully acknowledged.

References