AOBPreview originally published online on April 29, 2008
Annals of Botany 2008 101(9):1421-1432; doi:10.1093/aob/mcn054
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Mechanical Stimuli Regulate the Allocation of Biomass in Trees: Demonstration with Young Prunus avium Trees
1 INRA, UMR PIAF- Université Blaise Pascal, 234 avenue du Brézet, 63100 Clermont-Ferrand, France
2 INRA, UMR SYSTEM, Equipe d'Agroforesterie, Bâtiment 27, 2, Place Viala, 34060 Montpellier Cedex 1, France
* For correspondence. E-mail coutand{at}clermont.inra.fr
Received: 24 January 2008 Returned for revision: 25 February 2008 Accepted: 18 March 2008 Published electronically: 1 May 2008
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIAL AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
Background and Aims: Plastic tree-shelters are increasingly used to protect tree seedlings against browsing animals and herbicide drifts. The biomass allocation in young seedlings of deciduous trees is highly disturbed by common plastic tree-shelters, resulting in poor root systems and reduced diameter growth of the trunk. The shelters have been improved by creating chimney-effect ventilation with holes drilled at the bottom, resulting in stimulated trunk diameter growth, but the root deficit has remained unchanged. An experiment was set up to elucidate the mechanisms behind the poor root growth of sheltered Prunus avium trees.
Methods: Tree seedlings were grown either in natural windy conditions or in tree-shelters. Mechanical wind stimuli were suppressed in ten unsheltered trees by staking. Mechanical stimuli (bending) of the stem were applied in ten sheltered trees using an original mechanical device.
Key Results: Sheltered trees suffered from poor root growth, but sheltered bent trees largely recovered, showing that mechano-sensing is an important mechanism governing C allocation and the shoot–root balance. The use of a few artificial mechanical stimuli increased the biomass allocation towards the roots, as did natural wind sway. It was demonstrated that there was an acclimation of plants to the imposed strain.
Conclusions: This study suggests that if mechanical stimuli are used to control plant growth, they should be applied at low frequency in order to be most effective. The impact on the functional equilibrium hypothesis that is used in many tree growth models is discussed. The consequence of the lack of mechanical stimuli should be incorporated in tree growth models when applied to environments protected from the wind (e.g. greenhouses, dense forests).
Key words: Prunus avium, growth, mechanical stress, bending, biomass, shoot/root ratio, wind, shelter
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIAL AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
Plastic tree-shelters are commonly used in forestry and agroforestry plantations to protect tree seedlings against various hazards such as browsing animals or herbicide drifts (Dupraz, 1997b). This type of shelter has the advantage of being light, easy to carry and easy to install. These tree-shelters induce a greenhouse effect and modify tree growth by altering the microclimate within the shelter (Bergez, 1993). A reduction in root and stem diameter growth has been observed (Dupraz et al., 1993; Svihra et al., 1996; Dupraz, 1997b). While earlier studies focused on temperature and light modifications to explain the distorted growth of trees inside tree-shelters, more detailed studies have proved that CO2 availability is the most crucial aspect (Dupraz and Bergez, 1999). The ventilation rate resulting from free convection through the top of the shelter was too low to supply the tree with enough CO2, and resulted in a reduced assimilation rate of the tree (Dupraz and Bergez, 1999). This impact was so pronounced that in some species such as walnut, the final biomass after 1 year of growth inside a shelter was less than the initial biomass at planting, despite a similar or improved growth in height (Dupraz, 1997a). Ventilated shelters with holes at their base were suggested in order to overcome this problem (Bergez and Dupraz, 2000) and resulted in normal assimilation rates and total biomass of the trees. However, the shoot–root ratio remained unbalanced; but the explanation for the reduction of the root system has not proved to be straightforward. One hypothesis is that the abnormal biomass allocation results from the lack of tree movement within the shelter.
Several studies have demonstrated that mechanical stimuli (generated by wind sway, for example) lead to a reduction in height and to an increase in stem radial growth (Jacobs, 1954; Mäkelä and Sievanen, 1992). This phenomenon has been called thigmomorphogenesis (Boyer, 1967; Jaffe, 1973). The lack of mechanical stimuli (in staked trees for example) leads to less-tapered trees (Larson, 1965). In tall pines, it was demonstrated that the reduction in wind sway by tethering induced an increase in height growth (Meng et al., 2006).
Many studies of thigmomorphogenesis have also been conducted on herbaceous species and have assessed the changes in above-ground plant dimensions with different mechanical stimuli (see Biddington, 1985, for a review on herbaceous species). Some of these studies were dedicated to comparing the effect of wind sway with artificial mechanical stimuli such as bending. For example Latimer et al. (1986) compared the effects of natural and artificial mechanical stimuli on eggplant stems, and found that artificial bending and wind resulted in similar effects on stem dimensions. Smith and Ennos (2003) reported a study on sunflower in which they assessed if wind had an effect on gaseous exchange and/or a ‘thigmo’-mechanical effect. Both studies concluded that the effect of wind is firstly thigmomorphogenetic.
In addition to their effect on stem dimensions, mechanical stimuli also affect biomass allocation between roots and shoots. This effect has been mainly studied on herbaceous species (Biddington and Dearman, 1987; Gartner, 1994; Niklas, 1998).
In addition, mechanical stimuli can also have an impact on rooting patterns. Goodman and Ennos (1998) demonstrated that the shape of root systems was modulated by wind. In Zea mays and Helianthus annuus lateral roots perpendicular to the direction of the bending plane were less developed than roots in the direction of the bending plane. In maize, the leeward roots were thicker and more numerous than the windward roots. In contrast, wind increased the number of roots and the root thickness of the leeward and windward sides in sunflower (Goodman and Ennos, 1998). These differences indicate that the response of root systems can vary between species. With regard to trees, a recent study on oak seedlings demonstrated that wind-loaded plants produced twice as many roots that were longer than those of the control plants, and an asymmetric root system with windward roots being significantly more numerous and longer than leeward roots (Tamasi et al., 2005).
Thigmomorphogenesis takes place where the mechanical stimulus is applied, but also at a distance: a thigmomorphogenetical response is observed on plant parts that are not loaded. Bending in the basal part of the stem leads to a growth response in the apical part (Coutand et al., 2000), mechanical signals generated by soil compaction have been shown to reduce shoot size (Young et al., 1997; Masle, 1998), and pricking on cotyledons reduced the length of the hypocotyl (Desbiez et al., 1981).
In the case of bending, the mechanical state of a structure is described by kinematic variables such as displacement (axis scale); curvature (cross-section scale); strain (local scale) and by static variables such as force (axis scale), bending moment (cross-section scale) and stress (local scale). It has, however, been demonstrated that plants perceive their state of strain, and not forces or stresses (Coutand and Moulia, 2000). In the case of growth in length, it has been demonstrated that the bent part of the plant perceives strain and that these local perceptions are summed up in order to produce a signal that modifies the growth of the elongation zone. In the case of radial growth of roots (Goodman and Ennos, 1996, 1998, on sunflower and maize) or stems (C. Coutand, on walnut and poplar, unpubl. res.) mechano-perception seems to be more local.
At a cellular or subcellular level, our knowledge is expanding rapidly with the identification of genes implicated in mechano-perception (e.g. Lee et al., 2005) and with models that describe the links between the cell wall and plasma membrane deformation (see Fasano et al., 2002, and Telewski, 2006 for an excellent review on the subject).
Despite this progression at the molecular level, the regulation of thigmomorphogenesis remains unclear, and there are very few physiological studies providing quantitative data on growth and mechanical stimuli. Currently, there are no accurate strain threshold values, no quantification of a refractory period between two stimuli, nor evidence of some attenuation of the perceptive-system sensitivity due to acclimation phenomena.
The first aim of this work was to test whether the reduction in mechanical stimuli induced by a shelter may explain the shift in biomass allocation to the tree shoots. A factorial experiment was set up in order to de-correlate the microclimatic effect of the shelter from the ‘mechanical’ effects. The set-up of a factorial experiment led to an innovative device being designed that allows mechanical stimuli to be applied to trees inside the shelter. The work was conducted with young wild cherry trees by monitoring tree growth (in terms of stem elongation and stem diameter) on a weekly basis and by measuring the final tree biomass after the growing season. The second aim of this work was to check whether some quantitative relationships could be found between the mechanical stimuli induced by bending and the growth response of the shoots, and if some acclimation phenomena of trees to mechanical stimuli occurred.
|
MATERIAL AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIAL AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
Trees
One-year-old Prunus avium (‘Monteil’) were planted in February 2001 in 30-L individual containers filled with ‘Limagne’ soil. Limagne soil is a dark, Luvisol composed of clay (46·5 %), loam (26·6 %), thin sand (14·8 %), coarse sand (8 %) and 4·1 % of organic matter with a pH of about 8·2. Water was supplied three times a day by an automatic watering device, avoiding any water stress to the trees.
Treatments
The experiment was conducted outside. Four treatments were compared; each treatment was applied to ten trees. The four treatments consisted of: trees free to sway in the wind, (no shelter and wind, NSW); trees in a plastic shelter (shelter, S); trees in a plastic shelter but bent by a mechanical device (shelter + bending, SB; Fig. 1A); and trees progressively staked with a wooden post during the growing season in order to reduce trunk movement (no shelter + stake, NSSt; Table 1). The sheltered trees (S and SB) were protected by a green-coloured, translucent, circular, twin-walled polypropylene tube (Tubex, Aberdare, England), 2·1 m tall and 0·11 m wide with ventilation holes at the bottom of the shelter (‘Tubex équilibre’ Tubex®), and were prevented from any movement because of the shelter. The trunk of the bent trees (SB) was set against two fixed points separated by a distance of 80 cm. Mid-distance between these two fixed points, a moving arm linked to a pressurised-air dash-pot (Fig. 1B) enabled the trunk to be bent. Eight bending movements were applied automatically each day at 3-h intervals controlled by an electronic clock. The maximal bent state was reached almost instantaneously and the tree stayed in the bent state for 1 min, then the force was removed. The trunk of the staked trees (NSSt) was prevented from moving but the branches were free to move and the leaves were free to rustle in the wind. Stakes were put on the north side of the trees in order to reduce the shading effect. The variability of the tree seedlings (stem height and diameter) was controlled at planting time. Similar distributions in terms of stem size for the different treatments were obtained by stratification.
|
|
Growth measurements
The height of the annual shoot was measured weekly with a ruler (precision ± 0·5 cm). In cases of apical bud death (only four trees), the height was monitored with the new leader. The diameter of each tree was measured weekly on each trunk with a Vernier caliper at 15 and 65 cm from the collar (collar refers to the stem base in this paper). At the end of the growing season, the trees were uprooted. Leaves, trunk, branches and the annual shoot were separated, dried for 3 d at 103 °C and weighted after 1 d of rehydration in the ambient air of the laboratory. The root system was washed, dried and weighted following the same procedure as the shoots.
The biomass on the day of planting was estimated for all seedlings using an allometric relationship obtained from an additional set of ten trees: knowing the diameter of the trunk at three locations, the trunk volume was computed as the sum of two truncated cones. Trunks were separated from the root system, weighed, dried and then weighed again; this enabled the percentage of water in the fresh biomass to be estimated. The relationship between trunk biomass and fresh trunk volume provided the following linear equation:
|
|
For planted trees, the dimensions of the trunk were measured. The fresh and dry trunk biomass was then computed using this allometric relationship. The total fresh biomass was measured. The biomass of the root system was then computed as the difference between the total biomass and the trunk biomass. The net production of biomass during the growing season was obtained from the difference between the biomass at the end of the growing season and at planting time.
Response of radial growth to mechanical bending
In order to quantify the tree growth response to mechanical bending, and the state of tree deformation, controls were required. In natural conditions plants are always subjected to mechanical stimuli and therefore the ‘control plants’ were those that were free to move; NSW was thus considered as the control treatment.
Quantification of plant responses
The thigmomorphogenetical response of trees was computed as the difference in growth between stimulated and control trees. It was not considered relevant to pair each plant of the SB treatment with a plant from the S treatment or NSB treatment, since no significant relationship was found between the initial stem diameter and the gain in stem diameter in the NSW treatment, either at weekly intervals or over the entire growing season. At an individual level, the growth response (G) was computed as the difference between the gain in diameter of each SB tree and the average gain in diameter of the NSW treatment (natural control). In addition, the growth response of each SB tree was computed as the difference between the gain in diameter of each SB tree and the average gain in diameter of S trees. Secondly, an ‘average plant’ was also considered for the SB treatment.
At the tree level, it can be written:
|
| (1) |
At the treatment level, it can be written:
|
| (2) |
The same procedure was used for the growth response in terms of stem length. Within the shelter, bending was applied on the basal part of the stem but trees also experienced bending on the stem above the bending system: the displacement of the distal part of the stem was blocked by the shelter so a counter-curvature took place in the distal part. As thigmomorphogenesis is known to have effects both locally and also at a distance, the length growth might have been affected by the sum of the bending and counter-bending signals (Fig. 2): the basal part of the stem might also be subjected to a thigmomorphogenetical signal generated by the counter-bending of the upper part of the stem. The data obtained did not allow the mechanical state within the distal part to be quantified and, as a consequence, in this paper only the signal coming from the bent basal part is considered.
|
Estimated quantification of the strain state due to artificial bending
The applied bending was controlled in terms of displacement: the middle point of the stem was laterally displaced by 4 cm. To quantify the strain state imposed by artificial bending, the free software RDM Le Mans 6·16 was used (http://iut.univ-lemans.fr/ydlogi/rdm_version_6.html). The frame toolbox (‘ossature’) rather than the beam toolbox (‘poutre’) was used because it enables computations with tapered beams, which is not the case with the ‘poutres’ toolbox. The software enables the computation of mechanical variables describing an elastic structure's mechanical state in tapered beams with chosen imposed limiting conditions and assuming small strains. The stems were fully elastic in the range of loadings that were applied and recovered their initial shape after the piston was put back. In this case, the boundary conditions were fixed points located at 5 and 85 cm from the collar (Fig. 2B). Knowing the stem geometry (diameter variation along the stem), the imposed displacement and the longitudinal Young's modulus of the green wood (ELgreen), the software computes the bending moment. The taper of each stem was taken into account by two linear interpolations assuming that the trunk was constituted of two truncated cones: diameter values at 15 cm and 65 cm from the stem base were available and the diameter under the apex was set at 0·5 cm.
Knowing the bending moment enabled the longitudinal deformation to be computed at a distance h from the stem base:
|
| (3) |
The value of ELgreen(h) was estimated from the literature; a value of 10 200 MPa was assumed for the elasticity modulus of wild cherry at 12 % humidity (Benoit and Dirol, 2000). The elasticity modulus of green wood is then given by:
|
| (4) |
where H is the relative humidity; EL is the elasticity modulus at 12 % humidity (Jodin, 1994).
Conceptual framework to link the plant response and the applied level of strain
It is not possible to discuss the growth response of a plant to successive imposed strains without considering the regulation of the thigmomorphogenetic process. To our knowledge, most work on thigmomorphogenesis has used successive mechanical stimuli, but there is no work available that has attempted to integrate the effect of successive mechanical stimuli. A conceptual framework is therefore suggested, as follows.
- If the plant has no memory of the mechanical stimuli it has experienced (i.e. no acclimation occurs), and no refractory period, all the applied strains are perceived and lead to a growth response. There should be a linear relationship between the imposed strain and the response. This relationship can be tested both at weekly intervals (the measurement frequency) and on a growing-season time scale. The growth response was documented with weekly diameter measurements; the level of imposed strain had to be computed using the same time interval. It was assumed that the stem diameter was constant on a daily basis, but changes from one week to the next were taken into account by the measured increase in stem diameter.
- The level of imposed strain per day is given by the product of the level of imposed strain and the frequency of bending (8 d–1). The level of imposed strain per week is the sum of the successive levels of strain per day. The level of imposed strain on the growing-season scale is the sum of the strain per week.
- If the plant has a memory of the previous mechanical stimuli (Desbiez et al., 1984), there should be a non-linear relationship between the imposed strain and the response, as suggested on tomato stem for elongation (Coutand and Moulia, 2000).
- If the plant has a memory of the previous mechanical stimuli and an increase in the strain threshold, there should be a relationship between the effective part of the imposed strain and the growth response. With the data from this study the hypothesis of an increase in the strain threshold at the level of the previously applied strain was tested. The effective strain at time t, straineff(t), was computed as
|
| (5) |
Relationship between the imposed strain state and tree growth response
First, the pattern of the imposed strain state with time was documented. It was then checked to see if there were statistically significant relationships between the imposed strain and the tree responses. Finally, the kinetics of the response with time was studied to check whether some acclimation processes could be discerned.
Statistical tests
The experimental design was balanced (same number of individuals per treatment), and ANOVA procedures and Student's t-tests were used to compare growth results (P < 0·05). Statistical tests relating to biomass data, and particularly ratios, normally require the use of transformed variables. However, as the ratios did not differ very much from 0·5 ± 0·2, the classical Student's t-test could be used for these analyses (A. Lacointe, UMR PIAF, Clermont-Ferrand, pers. com.).
The relationship between the imposed strain and the tree response in terms of cambial growth had to be checked carefully as both quantities involve the stem diameter in computations: the experimental design had to be checked retrospectively. No significant relationship was found between the initial diameter and the seasonal increment in diameter for trees of the SB treatment at 15 cm from the stem base. In contrast, a significant relationship was found between the initial diameter and the seasonal increment in stem diameter at 65 cm from the stem base (P = 0·0011), with the diameter explaining 12·9 % of the variability of the increment in diameter.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIAL AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
Kinetics of longitudinal and radial growth
The total elongation of the annual shoot in sheltered trees was impressive: 2- to 3-fold longer than that of unsheltered trees (Fig. 3A). Artificial mechanical stimuli imposed inside the shelter significantly reduced the final elongation (P < 0·0001); the impact on elongation was noticeable as early as only 2 weeks after the start of artificial bending (Fig. 3A). Staked trees (NSSt) had a significantly longer elongation of the annual shoot compared to NSW trees, and shorter compared to sheltered trees (S).
|
Radial growth was significantly increased for the bent trees (Fig. 3B). Until 20 July, radial growth at the stem base was the same for all sheltered/staked trees (S, SB and NSSt treatments); however, 2 weeks after the application of bending, the cambial growth of the trees of the SB treatment suddenly increased at the stem base (Fig. 3B). From 18 August the difference in diameter between the SB trees and the control trees disappeared (P = 0·13). This effect was similar at 65 cm from the stem base (data not shown) and even more pronounced at the end of the experiment when trees of the SB treatment exhibited a significantly higher gain in diameter than the control trees (P < 0·05). Throughout the whole growing season, motionless trees (S and NSSt) exhibited no differences in radial growth at the stem base.
The consequences of artificial bending appeared even more clearly when examining the differential of radial growth of the SB treatment compared with the NSW treatment. At the stem base, the differential exhibited negative values before bending was applied, but thereafter these values became less and less negative with time, indicating that bending was increasing the radial growth of SB trees. The differential values remained negative but tended toward zero (Fig. 4A), indicating that the artificial bending was sufficient to lead to radial growth similar to that generated by wind. When comparing the differential of radial growth of the SB treatment with the NSW treatment at 65 cm from the stem base, the differential values became more and more negative before bending was applied and were then more positive, indicating that artificial bending led to radial growth superior to the radial growth rate of control trees (Fig. 4B).
|
It is interesting to look at the effect of mechanical stimuli versus time by considering the kinematics of the apical shoot growth of trees during the experiment. It appears that the trees followed a pattern that is classical for most temperate hardwood species: a first flush of growth at the end of May and a second flush around mid-July (Fig. 5). The comparison of S and SB growth rates revealed three phases: the installation of the bending system firstly induced a transient reduction of the elongation rate. Then SB trees recovered to an elongation rate close to that of the S trees. Finally, artificial bending lead to a sudden decrease in the elongation rate. The mechanical stress due to the installation of the bending system was sufficient to affect the elongation rate for 2 weeks (Fig. 5) but did not significantly affect the radial growth (data not shown).
|
Effects of the treatments on biomass and biomass partitioning
Both the final total biomass and the biomass produced during the growing season did not differ between treatments [P = 0·7154 and 0·39, respectively]. At the date of planting, there was no significant difference [P > 0·07] in the partition of biomass between the shoots and roots in the four sets of trees, with 45 % of the biomass in the roots. After treatment, 60 %, 45 %, 37 % and 56 % of the biomass was allocated to the root system for NSW, SB, S and NSSt trees, respectively (Fig. 6).
|
The difference in biomass partitioning was significant between treatments, [P = 0·001] except between the NSW and NSSt treatments (however, the probability was low: P = 0·13). Sheltered trees had a significantly reduced root fraction compared with trees mechanically stimulated within the shelter. Mechanical stimuli inside the shelter significantly increased the allocation of biomass towards the roots, but did not allow the trees to attain a root/shoot ratio close to that of the control trees (Fig. 6).
Consequences of the lack of mechanical stimuli on tree behaviour
The consequences of the lack of movement on changes in biomass partitioning were important in terms of the growth habit of the trees. When the shelters of the S treatment were removed, the height/diameter ratio of the trees was such that they were unable to support themselves and they buckled under their own weight (data not shown).
Quantitative links between shoot dimensions and imposed strain state
The level of applied strain ranged from 0·5 to 6 % at the stem base, with an average value of 4·4 ± 0·8 %. The level of applied strain ranged from 0·3 to 2·5 % at 65 cm from the stem base, with an average value of 1·8 ± 0·3 %. The imposed strain increased with time (Fig. 7), and was correlated with the radius of the stem (Pearson's coefficient: 0·89 at 65 cm, and 0·52 at 15 cm from the stem base; data not shown). The longitudinal strain at the periphery of the stem is the product of the curvature and the stem radius. However, data analysis showed that the strain was highly correlated with the radius (see Material and Methods), indicating that the curvature did not vary very much throughout the experiment.
|
In order to analyse the relationship between growth and strain, data sets from 30 July were removed: at that time growth rates were extremely low, resulting in a very poor precision of the estimation of growth response. No significant relationships were found between imposed strain and growth responses either weekly or over the growing season at 15 cm from the stem base. There was a low but significant relationship between the response and the imposed strain at 65 cm from the stem base (Fig. 8), which only explains 20 % of the variability, which was not much higher than the percentage variability explained by the stem radius alone. When the data sets are considered with regards to the sub-sets of points per date, it suggests that there was a decrease in response with time, which might be indicative of an acclimation process in the trees (Fig. 8). In order to check this trend, the growth response kinetics were studied.
|
Growth response kinetics
Considering an average plant, and taking the NSW trees as the control treatment, the growth response increased during the first 2 weeks after application of bending and then exhibited a global tendency to decrease. This was found for the response in terms of both height (Fig. 9A), and cambial growth at the stem base and at 65 cm from the stem base (Fig. 9B, C).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIAL AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
Effect of mechanical stimuli on the shoot/root ratio
In this experiment, biomass production did not differ between treatments but the shoot/root ratio of trees in tree-shelters (although ventilated) was highly distorted, as found by Dupraz (1997b) for ten different tree species, including conifer and broad-leaved species, and later confirmed by Bergez and Dupraz (2000) for wild cherry trees. This showed that assimilate partitioning is not only influenced by the assimilation process intensity, but more probably by a variable that is not connected to ventilation. Bergez and Dupraz (2000) proposed three hypotheses for the mechanism responsible: light intensity, light quality or the movement of the stem and leaves induced by the wind. In greenhouse experiments, seedlings that were shaken to simulate wind effects had lower height growth, greater diameter growth and a shorter period of extension growth compared with trees that remained motionless (Neel and Harris, 1971, 1972; Mitchell and Myers, 1995; Osler et al., 1996).
The comparison of S and SB trees demonstrates that the lack of movement induced by the shelter was involved in the distortion of the biomass allocation for sheltered trees. The effects of mechanical stimuli were impressive, considering the stimuli were applied very late (only from mid-July onwards) and lasted only approx. 6 weeks.
It was surprising to observe no difference in biomass partitioning between NSW and NSSt trees. This suggests that the reduction in stem movement in NSSt trees was compensated for by another thigmomorphogenetical signal. The leaves, that were free to move in both the NSW and NSSt trees, may be involved, but the longitudinal strain on the stem resulting from the movement of thin leaves was probably very low. If this hypothesis is confirmed, it would suggest that the leaves are extremely sensitive from a thigmomorphogenetical point of view and able to produce a signal that is sufficient to induce thigmomorphogenetical responses within trunks and roots.
Kinetics of plant response
The effects of artificial mechanical stimuli were observed on tree growth 2 weeks after the start of the treatment. This delay is in accordance with what was found for a dicotyledonous plant (sunflower) and slightly longer compared with a monocotyledonous plant (maize; Goodman and Ennos, 1996). Over short time scales, plants develop a growth response to mechanical stimuli very quickly. In experiments with a higher frequency of measurements, the time necessary for the plant to react in terms of growth were found to be far smaller: 3–8 min for tomato stems (Coutand et al., 2000), sunflower stems (Peacock and Berg, 1994), wheat and barley (Young et al., 1997), maize (Jaffe et al., 1985) and bean (Jaffe, 1976). These results seem discordant with the results found in this study, but the apparent discrepancy may be attributed to differences in response between herbaceous and woody plants (it would take more time for a woody plant to respond to an applied mechanical stimulus), or to the combination of measurement accuracy and frequency used in this study. The latter hypothesis is more probable: young poplar trees subjected to bending also showed very rapidly induced modifications in radial growth (C. Coutand, unpubl. res.). The accurate detection of the time at which growth begins to be modified is a difficult experimental issue. Usually, the short times detected in the above-mentioned studies correspond to the growth-monitoring frequency, so the actual delay before the plant begins to react is certainly even shorter, and the effect is probably almost instantaneous.
Effect of mechanical stimuli on stem dimensions
The effects of mechanical stimuli on stem dimensions were quantitatively important: mechanical stimuli of sheltered trees reduced the length of the annual shoot developed after the start of the stimuli by about 80 % and increased the radial growth at the stem base by about 30 % in comparison with sheltered (S) trees. Staking resulted in similar differences in elongation between staked trees and trees free to move in the wind. The NSSt treatment continually reduced the cambial growth in comparison with the NSW treatment. In both treatments the leaves could rustle, and the main difference between these treatments was the movement of the stem.
The length of the annual shoot and the diameter of NSW and SB trees did not differ significantly at the end of the growing season. This demonstrates that the very low frequency of artificial bending stimuli (one bending every 3 h) was sufficient to match the impact of the natural wind. Moreover, artificial bending was applied for only 6 weeks, starting after 15 July, whereas the wind was present throughout the whole experiment (from 15 May until the end of August). This demonstrates a powerful effect of the applied artificial bending on tree stem growth: stem elongation was almost stopped, and the low bending frequency allowed only a very small elongation between two successive bending motions.
It is, however, unclear how the plant regulates its perception when it experiences successive mechanical stimuli. During the experiment, trees with no shelter were continuously submitted to wind-imposed mechanical stimuli. If the relationship between mechanical stress and growth that was measured on sheltered trees is extrapolated to the NSW trees, they would not grow in length. Four non-exclusive hypotheses can be formulated: (1) some mechanical stimuli induced by the wind may not be perceived because they do not induce sufficient strain; (2) there may be an effect of the variations in strain rate on perception – for a given level of applied strain, plants may react differently if the imposed strain rates vary; (3) plants may exhibit a refractory period during which they ignore the stimuli; and (4) plants may acclimate to the imposed strain. The first two hypotheses have been demonstrated experimentally with work conducted on tomato stem elongation by Coutand and Moulia (2000) where a threshold of strain was demonstrated but not precisely quantified, and by Coutand (1999) where very low rates of mechanical stimuli induced a very limited or no response. The third hypothesis could not be tested with the experimental design used in this study. The last hypothesis was demonstrated as the tree response decreased while the imposed strain increased.
Effect of mechanical stimuli on biomass partitioning and its consequences from practical and fundamental points of view
Mechanical stimuli did not affect the total tree biomass, but modified the biomass allocation (1) within the aerial part (between primary and secondary growth) and (2) between the above-ground and below-ground parts of the tree. Surprisingly, staked and free-to-move trees exhibited no statistically significant differences in their root/shoot ratios. This is at odds with the stimulated height growth observed, but consistent with a similar pattern of diameter growth. Two hypotheses can be suggested. Staking trees during their growth may have induced permanent mechanical deformations with an impact on stem elongation but not on stem diameter growth or biomass allocation to the roots. This would suggest that dynamic and static strain have opposite effects on plant development. An alternative hypothesis suggests that the movements of leaves and/or of the few branches were enough to induce effective mechanical signals. Inside the tree-shelters of the S treatment, leaves remained still and never rustled, and branches never moved.
The results demonstrate that applying artificial mechanical stimuli is a way to modulate the allocation of plant biomass, and so to control the stem dimensions and the shoot/root ratio. In this study, the application of mechanical stimuli within shelters may have compensated for the deficiency of movement induced by the shelter and favoured stem diameter and root growth.
The results clearly demonstrate that mechanical stimuli greatly influence biomass partitioning, which raises questions from a more theoretical point of view. The determinism of biomass partitioning is still rather unclear. In functional–structural tree models (FSTMs) the regulation of biomass by mechanical stimuli is, in most cases, not taken into account (Mäkelä, 1999, 2003; Lacointe, 2000). With regard to models of tree growth, Lacointe (2000) distinguished four model types: in the first type, empirical allocation coefficients are used (e.g. Leriche et al., 2001); in the second type, biomass partitioning relies on dynamic optimum relations (e.g. pipe-model, constant root/foliage ratio; see Mäkelä, 2003, for a review); in the third type, analogies with electrical schemes are used; the fourth type is more mechanistic – biomass partitioning is determined by the sink strength of organs. The sink strength is a theoretical concept defined as the potential growth an organ would achieve in unlimiting growth conditions. The quantity of biomass allocated to a sink is usually determined using empirical laws for the sink strength: for example, when the branch biomass is derived from branch dimensions (Grote and Pretzsch, 2002; Lacointe et al., 2002) the sink strength is determined by the volume of meristem and its respiration (Bidel et al., 2000). The sink strength can be modulated by environmental factors (Lacointe, 2000) and by the distances between sources and sinks: sources preferentially supply the closest sinks (Bidel et al., 2000; Lacointe, 2000).
The results of the current study suggest that the sensing of mechanical stimuli should also be included as one of the factors involved in the determinism of biomass allocation. Until now, the regulation of biomass partitioning based on mechanical considerations has been almost absent from FSTM models (Lacointe, 2000). The few mechanical models of tree growth are not linked to assimilate partitioning; they include mechanical constraints in terms of mechanical safety, regulating the dimensions of tree axes to ensure that the tree does not buckle under its self-weight. In these studies, branch or stem tapering follow a power law with the exponent being equal to 1·5 (McMahon and Kronauer, 1976; cited by Lacointe, 2000). A recent study on the critical height of trees has been carried out on juvenile trees in tropical rain forests, taking into account mass distribution and trunk-tapering that is different from 1·5 (Jaouen et al., 2007). It was demonstrated that the computation of buckling height using the classical formula (exponent 1·5) leads to an under-estimation of the critical height. The results of this study could be used in FSTM models where the mechanical optimum is taken as a growth rule in order to partition assimilates by improving the calculation of the critical height with a small number of parameters. To our knowledge, Ford et al. (1990) are the only authors who have used an exponent different from 1·5 to partition assimilates between elongation and cambial growth in coniferous branches, so that the branches exhibited a certain deflection profile.
In AMAPmeca (Fourcaud and Blaise, 2003; Fourcaud and Lacq, 2003) the shape variations of growing branches have been taken into account in order to simulate more realistic axis shapes (S-shaped branches are common in nature) through tree development, including gravitropic reactions of trees. This is an extension of a model, developed on a young pine (Fournier et al., 1994), to trees exhibiting ramifications. However, none of these models has linked the mechanics of axes with assimilate partitioning.
The experimental curves of tree growth subjected to mechanical stimuli (thigmomorphogenesis studies) could be used to introduce the impact of these stimuli on assimilate partition in trees. However, the curves presented in this study apply only to cherry trees under certain conditions of mechanical stimulation. An alternative option would be to modify the sink strength using mechanical stimuli. This would require establishing action laws, i.e. quantitative relationships between, for example, the mechanical state of the plant and the modifications of biomass allocation between the shoots and the roots, or between the mechanical strain and the length or diameter increase of axes after a single bending (Coutand and Moulia, 2000). Such action laws should also include the case where several mechanical stimuli are applied to the plant, which highlights the importance of studying the regulation of the thigmomorphogenesis process. As stated in the Introduction, shoots and roots are sensitive to mechanical strain, but until now a general action law has not yet been established. If such a relationship exists it could then be used to modulate the sink strength.
Quantitative links between shoot response and mechanical stimuli
The imposed strain in this experiment was quantified, which is rarely found in thigmomorphogenesis studies. It demonstrated that small strains (range of magnitude from 1 to 5 %) applied at a low frequency (eight bending movements per day) can lead to impressive effects on plant growth. For the growth response at 65 cm from the stem base, the variability in curvature was low and the relationship between the response and the imposed strain leads to a very close relationship between the response and the stem radius. In order to see if a relationship exists between the imposed strain and the response, it would be necessary to control the curvature in order to introduce a higher variability within the data set.
The poor or negative correlation between the imposed strain and the cambial response also suggests an acclimation of trees with time and/or strain history. This is confirmed by the kinetics of the cambial response of an average plant. The decrease in growth response appeared only 3 weeks after bending was first applied, suggesting that an acclimation process took place very rapidly. The response globally decreased with time, but sometimes exhibited isolated peaks. The sudden decrease in the growth response around the end of July and the beginning of August may be a measurement artefact: at this time of the season growth is very limited, so measurement precision was low. If the peaks are not a measurement artefact, they may indicate the sensitivity of the plant to its growth rate. Unfortunately, the data did not allow this hypothesis to be checked because the variation in diameter is used for the computation of the plant response and the plant growth rate. To elucidate possible interactions between successive bending, it would be necessary to compare several bending frequencies, including a single-bending treatment.
In this experiment it was shown that the effect of the installation of the bending system affected tree growth for 2 weeks. In practical terms, if mechanical loading is used to control tree growth then bending trees every 2 weeks could possibly be sufficient, and more effective than higher bending frequencies.
Trees respond to the sum of several thigmomorphogenetical signals. The experiment here only allowed the verification of the effect of the thigmomorphogenetical signal coming from the bent part of the trunk. Establishing action laws, i.e. quantitative links between the tree growth response and the mechanical strain state of plants, would therefore require measuring the strain state of the different plant parts, i.e. to document the spatial and temporal strain fields and responses. The relationship between the movement (and strain state) of a living structure and the wind dynamics presents an intricate problem. Experimental and modelling tools have been developed in order to compute the strains experienced by a tree swaying in the wind (Y. Brunet, INRA, Bordeaux, France, unpubl. res.). The coupling of these two modelling approaches could therefore be considered. Interestingly, work by Py et al. (2006) has shown that there is a lock-in mechanism in the tree response to wind: the frequency of instabilities deviates from its expected value when approaching the natural frequency of the plant. In other words, at some particular wind frequencies, which are close to the plant's frequency, it has been demonstrated that the wind does not impose a swaying rhythm on the plant, but that the plant imposes a turbulence mode on the wind (Py et al., 2006). This result might, however, be limited to particular frequencies and may not be taken as a general rule; it also has to be checked for trees.
Conclusions
This work has documented the effect of mechanical stimuli on stem growth and biomass partitioning between roots and shoots on wild cherry tree seedlings. The impact of natural wind was compared with artificially reduced wind mechanical stimuli by staking or sheltering. Tree-shelters are widely used in forestry and agroforestry plantations, generating concern that sheltered trees may suffer from poor root development or unstable stem height/diameter ratios.
It was demonstrated that the deficit in root biomass and the reduced diameter induced by the use of individual shelters on young trees is explained by the absence of tree movement within the shelters. Surprisingly, few and limited artificial mechanical stimuli stimulated stem diameter and root growth while the total tree biomass remained unchanged. To date, the determinism of biomass allocation has remained rather controversial, but the functional equilibrium paradigm is widely accepted (Mäkelä, 1999), predicting that light, water or nutrient stresses shift biomass allocation to the most efficient parts of the plant to reduce the stress. This plasticity of the plant to adapt to stress is probably controlled genetically. It has been demonstrated here that mechano-sensing can play an important role in this determinism in some environments, such as protected locations where wind mechanical stimuli are reduced (e.g. greenhouses, shelters, understorey of dense forests, high density tree stands, urban trees close to buildings). Moreover, when trees are prevented from moving, it has been confirmed by this study that the biomass allocation can be modified in such a way that trees can no longer support their own weight and risk buckling under sudden strain. In contrast, sparsely planted trees face sustained wind stimulation, resulting in a very low shoot/root ratio and a high trunk taper (Mulia and Dupraz, 2006), making them very resistant to the risks of wind damage.
The survival of a tree depends on its ability to access resources (carbon, light, water) while bearing its own weight. This work demonstrates that perception of mechanical stimuli induced by movements is a way for plants to adjust their biomass partitioning.
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIAL AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
We greatly thank everyone from the PIAF research Unit for their help during root washing at the end of the experiment in the summer of 2001. We thank the TUBEX company for providing the tree-shelters used in the experiment. We also thank Dr E. Toussaint (UMR LaMI, Clermont-Ferrand, France) for suggesting the use of RDM Le Mans for strain computations and Dr A. Lacointe (UMR PIAF, Clermont-Ferrand) for helpful discussions on statistics. We are also grateful to Mrs H. Lamprell for correcting the English in the manuscript.
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIAL AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
-
Benoit Y, Dirol D. Le guide de reconnaissance des bois de France (2000) Paris: Eyrolles, CTBA edn.
Bergez JE. Influence de protections individuelles à effet de serre sur la croissance de jeunes arbres. Interprétations physiologiques et perspectives d'amélioration des protections (1993) France: USTL, Montpellier. Ph.D. thesis.
Bergez JE, Dupraz C. Effect of ventilation on growth of Prunus avium seedlings grown in treeshelters. Agricultural and Forest Meteorology (2000) 104:199–214.[CrossRef][Web of Science]
Biddington NL. A review of mechanically induced stress in plants. Scientific Horticulturae (1985) 36:12–20.
Biddington NL, Dearman AS. The effects of mechanically-induced stress and plant growth regulators on the growth of lettuce, cauliflower and bean (Phaseolus vulgaris L.) plants. Plant Growth Regulation (1987) 5:183–194.[CrossRef][Web of Science]
Bidel LPR, Pages L, Riviere LM, Pelloux G, Lorendeau JY. MassFlowDyn I: a carbon transport and partitioning model for root system architecture. Annals of Botany (2000) 85:869–886.
Boyer N. Modification de la croissance de la tige de Bryone (Bryonia dioica) à la suite d'irritations tactiles. Compte-rendu de l'Académie des Sciences de Paris (1967) 264:2114–2117.
Coutand C. Etude biomécanique de l'effet d'une flexion contrôlée sur la croissance primaire de la tige de tomate (Lycopersicon esculentum Mill.). (1999) Bordeaux, France: Université de Bordeaux I. Ph.D. Mecanique, Sciences du Bois.
Coutand C, Moulia B. Biomechanical study of the effect of a controlled bending on tomato stem elongation: local strain sensing and spatial integration of the signal. Journal of Experimental Botany (2000) 51:1825–1842.
Coutand C, Julien JL, Moulia B, Mauget JC, Guitard D. Biomechanical study of the effect of a controlled bending on tomato stem elongation: global mechanical analysis. Journal of Experimental Botany (2000) 51:1813–1824.
Desbiez MO, Boyer N, Gaspar T. Hypocotyl growth and peroxidases of Bidens pilosus. Effect of cotyledonary prickings and lithium pretreatment. Plant Physiology (1981) 68:41–43.
Desbiez MO, Kergosien Y, Champagnat P, Thellier M. Memorization and delayed expression of regulatory messages in plants. Planta (1984) 160:392–399.[CrossRef][Web of Science]
Dupraz C. Les protections de plants à effet de serre: amélioration de leur efficacité par aération optimisée et luminosité accrue. Revue Forestière Française (1997) a 49:519–530.
Dupraz C. Les protections de plants à effet de serre: ce qu'en pensent les arbres. Revue Forestière Française (1997) b 49:417–432.
Dupraz C, Bergez JE. Carbon dioxide limitation of the photosynthesis of Prunus avium L. seedlings inside an unventilated treeshelter. Forest and Ecology Management (1999) 119:89–97.[CrossRef]
Dupraz C, Guitton JL, Bergez JE, De Montard FX. Broadleaved tree plantation on pastures: the treeshelter issue. (1993) Proceedings of the 4th International Symposium, Windbreaks and Agroforestry. Viborg, Denmark: Hedeselskabet. 106–111.
Fasano JM, Massa GD, Gilroy S. Ionic signaling in plant responses to gravity and touch. Journal of Plant Growth Regulation (2002) 21:71–88.[Medline]
Ford R, Ford ED. Structure and basic equations of a simulator for branch growth in the Pinaceae. Journal of Theoretical Biology (1990) 146:1–13.[CrossRef][Web of Science]
Fourcaud T, Blaise F. Numerical modelling of shape regulation and growth stresses in trees. II. Implementation in the AMAPpara software and simulation of growth. Trees (2003) 17:31–39.[CrossRef]
Fourcaud T, Lac P. Numerical modelling of shape regulation and growth stresses in trees. I. An incremental static finite element formulation. Trees (2003) 17:23–30.[CrossRef]
Fournier M, Baillères H, Chanson B. Tree biomechanics: growth, cumulative prestresses, and reorientations. Biomimetics (1994) 2:229–251.
Gartner BL. Root biomechanics and whole-plant allocation patterns – responses of tomato plants to stem flexure. Journal of Experimental Botany (1994) 45:1647–1654.
Goodman AM, Ennos AR. A comparative study of the response of the roots and shoots of sunflower and maize to mechanical stimulation. Journal of Experimental Botany (1996) 47:1499–1507.
Goodman AM, Ennos AR. Responses of the root systems of sunflower and maize to unidirectional stem flexure. Annals of Botany (1998) 82:347–357.
Grote R, Pretzsch H. A model for individual tree development based on physiological processes. Plant Biology (2002) 4:167–180.[CrossRef]
Jacobs MR. The effect of wind sway on the form and development of Pinus radiata D. Don. Australian Journal of Botany (1954) 2:35–51.[Medline]
Jaffe M. Thigmomorphogenesis: the response of growth and development to mechanical stimulation with special reference to Bryonia dioica. Planta (1973) 114:143–157.[CrossRef][Web of Science]
Jaffe MJ. Thigmomorphogenesis: a detailed characterization of the response of beans (Phaseolus vulgaris L.) to mechanical stimulation. Zeitschrift fur Pflanzenphysiologie (1976) 77:437–453.[Web of Science]
Jaffe MJ, Wakefield AH, Telewski F, Gulley E, Biro R. Computer-assisted image analysis of plant growth, thigmomorphogenesis, and gravitropism. Plant Physiology (1985) 77:722–730.
Jaouen G, Alméras T, Coutand C, Fournier M. How to determine sapling buckling risk with only a few measurements. American Journal of Botany (2007) 94:1583–1593.
Jodin P. Le bois, matériau d'ingénierie (1994) Nancy, France. Arbolor eds.
Lacointe A. Carbon allocation among tree organs: a review of basic processes and representation in functional–structural tree models. Annales des Sciences Forestières (2000) 57:521–533.[CrossRef]
Lacointe A, Isebrands J, Host G. A new way to account for the effect of source–sink spatial relationships in whole plant carbon allocation models. Canadian Journal of Forest Research (2002) 32:1838–1848.[CrossRef]
Larson PR. Stem forms of young Larix as influenced by wind and pruning. Forest Science (1965) 11:412–424.
Latimer JG, Pappas T, Mitchell CA. Growth responses of eggplant and soybean seedlings to mechanical stress in greenhouse and outdoor environments. Journal of the American Society for Horticultural Sciences (1986) 111:694–698.
Lee D, Polisensky D H, Braam J. Genome-wide identification of touch- and darkness- regulated Arabidopsis genes: a focus on calmodulin-like and XTH genes. New Phytologist (2005) 165:429–444.[CrossRef][Web of Science][Medline]
Leriche H, LeRoux X, Gignoux J, Tuzet A, Fritz H, Abbadie L, Loreau M. Which functional processes control the short-term effect of grazing on net primary production in grasslands? Oecologia (2001) 129:114–124.[CrossRef][Web of Science]
Mäkelä A. Acclimation in dynamic models based on structural relationships. Functional Ecology (1999) 13:145–156.[CrossRef]
Mäkelä A. Process-based modelling of tree and stand growth: towards a hierarchical treatment of multiscales processes. Canandial Journal of Forest Research (2003) 33:398–409.[CrossRef]
Mäkelä A, Sievanen R. Height growth strategies in open-grown trees. Journal of Theoretical Biology (1992) 159:443–467.[CrossRef][Web of Science]
Masle J. Growth and stomatal responses of wheat seedlings to spatial and temporal variations in soil strength of bi-layered soils. Journal of Experimental Botany (1998) 49:1245–1257.
McMahon TA, Kronauer RE. Tree structures: deducting in the principal of mechanical design. Journal of Theoretical Biology (1976) 59:443–466.[CrossRef][Web of Science][Medline]
Meng SX, Lieffers VJ, Ried DEB, Rudnicki M, Sillins U, Jin M. Reducing stem bending increases the height growth of tall pines. Journal of Experimental Botany (2006) 57:3175–3182.
Mitchell CA, Myers PN. Mechanical stress regulation of plant growth and development. Horticultural Reviews (1995) 17:1–42.[Medline]
Mulia R, Dupraz C. Unusual fine root distributions of two deciduous tree species in southern France: what consequences for modelling of tree root dynamics? Plant and Soil (2006) 281:71–85.[CrossRef][Web of Science]
Neel PL, Harris RW. Motion-induced inhibition of elongation and induction of dormancy in Liquidambar. Science (1971) 173:58–59.
Neel PL, Harris RW. Tree seedling growth: effects of shaking. Science (1972) 175:918–919.
Niklas KJ. Effects of vibration on mechanical properties and biomass allocation pattern of Capsella bursa-pastoris (Cruciferae). Annals of Botany (1998) 82:147–156.
Osler GHR, West PW, Downes GM. Effects of bending stress on taper and growth of stems of young Eucalyptus regnans trees. Trees (1996) 10:239–246.[CrossRef]
Peacock K, Berg AR. Effect of mechanical stress on sunflower (Helianthus annuus L.) hypocotyl growth. Annals of Botany (1994) 74:661–666.
Py C, de Langre E, Moulia B. A frequency lock-in mechanism in the interaction between wind and crop canopies. Journal of Fluid Mechanics (2006) 568:425–449.[CrossRef][Web of Science]
Smith VC, Ennos AR. The effects of air flow and stem flexure on the mechanical and hydraulic properties of the stems of sunflowers Helianthus annuus L. Journal of Experimental Botany (2003) 54:845–849.
Svihra P, Burger D, Harris R. Treeshelter effect on root development of redwood trees. Journal of Arboriculture (1996) 22:174–179.
Tamasi E, Stokes A, Lasserre B, Danjon F, Berthier S, Fourcaud T, Chiatante D. Influence of wind loading on root system development and architecture in oak (Quercus robur L.) seedlings. Trees (2005) 19:374–384.[CrossRef]
Telewski FW. A unified hypothesis of mechanoperception in plants. American Journal of Botany (2006) 93:1466–1476.
Young IM, Montagu K, Conroy J, Bengough AG. Mechanical impedance of root growth directly reduces leaf elongation rates of cereals. New Phytologist (1997) 135:613–619.[CrossRef][Web of Science]
CiteULike 
Connotea 
Del.icio.us What's this?
Related articles in Ann Bot:
- ContentSnapshots
Ann Bot 2008 101: NP.[Extract] [Full Text]
This article has been cited by other articles:
|
|
 |
|
  L. Martin, N. Leblanc-Fournier, W. Azri, C. Lenne, C. Henry, C. Coutand, and J.-L. Julien Characterization and expression analysis under bending and other abiotic factors of PtaZFP2, a poplar gene encoding a Cys2/His2 zinc finger protein Tree Physiol, January 1, 2009; 29(1): 125 - 136. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||