Acta Biotheoretica

, Volume 60, Issue 1–2, pp 83–97 | Cite as

The Role of Calcium in the Recall of Stored Morphogenetic Information by Plants

  • Marie-Claire Verdus
  • Camille Ripoll
  • Vic Norris
  • Michel Thellier
Regular Article

Abstract

Flax seedlings grown in the absence of environmental stimuli, stresses and injuries do not form epidermal meristems in their hypocotyls. Such meristems do form when the stimuli are combined with a transient depletion of calcium. These stimuli include the “manipulation stimulus” resulting from transferring the seedlings from germination to growth conditions. If, after a stimulus, calcium depletion is delayed, meristem production is also delayed; in other words, the meristem-production instruction can be memorised. Memorisation includes both storage and recall of information. Here, we focus on information recall. We show that if the first transient calcium depletion is followed by a second transient depletion there is a new round of meristem production. We also show that if an excess of calcium follows calcium depletion, meristem production is blocked; but if the excess of calcium is in turn followed by another calcium depletion, again there is a new round of meristem production. The same stored information can thus be recalled repeatedly (at least twice). We describe a conceptual model that takes into account these findings.

Keywords

Abiotic stimuli Information storage Information recall Memory Calcium Meristems Flax Plant 

1 Introduction

Many authors have drawn attention to examples of the memorisation of signals in plants; for reviews, see e.g. Thellier et al. (2000), Trewavas (2003) and Ripoll et al. (2009). In most cases, memory has been considered to be evidenced when the perception of the first signal modified the way in which a plant responded to subsequent signals. In our approach to plant memory, we have restricted the concept to cases that involve both the storage of information following a stimulus and the recall of that information at a later time. Such a definition is consistent with the common use of the term “memory” in all sorts of systems ranging from pocket calculators and computers to animals and humans. The existence of the storage and recall of information has been shown in “hypocotyl-elongation inhibition” (Desbiez et al. 1987) or in symmetry-breaking in the growth of Bidens seedlings (Desbiez et al. 1991). Subsequently, a more convenient experimental system has been developed based on the “induction of meristem production” in the hypocotyl of flax seedlings (Verdus et al. 1997).

Flax seedlings are sensitive to many different abiotic stimuli. These stimuli include wind, drought, cold shock, irradiation by low intensity electromagnetic radiation or even just transplanting the seedlings from one medium to another (the “manipulation stimulus”). The seedlings respond to these stimuli by the production of epidermal meristems in the hypocotyl on the condition that they are also subjected to a transient depletion of calcium (Verdus et al. 1996, 1997; Tafforeau et al. 2002, 2004). When calcium depletion was delayed after stimulating the seedlings, the production of meristems was also delayed (Verdus et al. 1997). This means that, following a manipulation stimulus, seedlings are able to register and store “meristem-production” information but that they require a transient depletion of calcium to allow them to make use of the stored information—to recall it—for the actual induction of meristems. For brevity, we say that a STO function has been switched on (or activated) when the seedlings have stored the information corresponding to an abiotic stimulus and that a RCL function has been switched on (or activated) when the seedlings have been enabled to recall the stored information by being subjected to transient calcium depletion. In a preceding paper, we focussed on the calcium-dependent nature of the STO function (Verdus et al. 2007). Here, we focus on the RCL function with a view to build a conceptual model of the plant-memorization system.

2 Experimental Methods

In the experiments reported here, only one type of abiotic stimulus was employed: the manipulation stimulus. Flax seeds (Linum usitatissimum L., var Ariane) were allowed to germinate for 3 days on plastic grids fitted on top of growth vessels. The germinated seedlings were either left to grow on the same grid and the same growth vessel where they germinated (non-stimulated controls) or they were sampled and set in the meshes of another plastic grid fitted on top of another growth vessel (seedlings subjected to a manipulation stimulus). The standard nutrient medium, slightly modified from Homes et al. (1953), contained 2.33 mM Ca2+. Enriched (up to 6.99 mM Ca2+) and depleted media (zero mM Ca2+ supplied) were also prepared. In the calcium-depleted solutions, the Ca2+-concentration ([Ca2+]) was thus restricted to the unavoidable calcium contamination, ε (with ε = a few tenths μM); moreover, the concentration of nitrate was maintained equal to 11.58 mM by increasing the concentration of potassium nitrate. The ionic analyses were carried out by atomic absorption photometry. For meristem counting, the seedlings were dipped into 50% (v/v) aqueous ethanol until they were transparent enough for an easy counting of the epidermal meristems in the hypocotyls with a conventional light microscope. Each experimental point corresponds to the mean value of the numbers of meristems counted in 10 seedlings. For more detail about the experimental conditions, see Verdus et al. (1996, 1997, 2007).

3 Expression of the Data

Concerning the ionic analyses, preliminary experiments had shown that the total calcium content of the hypocotyls of flax seedlings grown in the standard solution increased with time. To characterise unambiguously, at each value of the time, t, the effect of decreasing or increasing the calcium concentration of the growth solution on the calcium content of the seedling tissues, we have used the ratio, R, of the calcium content of the experimental seedlings to that of the seedlings grown in the standard solution.

The data relative to meristem production has been represented by {t, n} graphs in which n is the mean number of meristems counted per seedling as a function of the time, t (days). When graphs corresponding to different experimental conditions have to be compared with one another, it is often convenient to characterize each graph by a single quantitative parameter. The “mean value”, N, is a good candidate for that. Using the trapezium method for the evaluation of the integral, I, of each distribution of experimental points {t, n}, N was calculated by
$$ N = I/\left( {t_{f} -t_{0} } \right) \, = \sum\limits_{j = 1}^{f} {\left( {t_{j} -t_{j-1} } \right) \times \left( {n_{j} + n_{j-1} } \right)/ 2\left( {t_{f} -t_{0} } \right)} $$
(1)
in which t0 is the last time when a zero number of meristems was counted, tf is the last time when meristems were counted, tj1 and tj are two successive time–values and nj-1 and nj are the corresponding mean numbers of meristems counted per seedling. Another advantage of using the mean value is that the effect of the fluctuations on the n values is minimized in the process of integration.

4 Experimental Data

4.1 Ionic Analyses

The R values in the hypocotyls of flax seedlings transferred from the standard ([Ca2+] = 2.33 mM) to a calcium-deprived ([Ca2+] = ε) solution (i) decreased rapidly to a value around 0.3 (i.e. approximately 30% of the content of the control seedlings) and then continued to decrease more slowly, and (ii) increased when the seedlings were returned to the standard solution (Fig. 1a), slowing eventually to reach an R value of only 0.8 after 3 days. Related experiments gave similar results and showed that at least 1 week in standard solution was needed for the calcium content of such seedlings to return to normal (data not shown). When the calcium depletion was only partial ([Ca2+] in the range of 0.58–1.75 mM), the decrease in the calcium in the seedlings was less pronounced than after complete calcium depletion, but the time-course of the return to normal calcium content was not appreciably modified (data not shown). When seedlings, previously grown in the standard solution ([Ca2+] = 2.33 mM), were transferred to a solution with a high calcium concentration (6.99 mM), the R value of the hypocotyls increased (with a significant fluctuation) up to an averaged value of approximately 1.5 after 3 days, i.e. on day 8 (Fig. 1b); then, when the plants were transferred to the calcium-depleted solution ([Ca2+] = ε), the decrease in the calcium content of the seedlings after 3 days (day 11) only reached an R value slightly above 1, and then on return to the standard solution remained roughly constant but with major fluctuations. Under calcium conditions identical to those in Fig. 1b (i.e. Ca concentrations of 6.99 mM followed by ε and 2.33 mM), the R values of the other ions (Mg2+, Na+ and K+) in the hypocotyls remained close to 1 (data not shown).
Fig. 1

Time-course of the relative calcium content, R, of the hypocotyl of flax seedlings grown under various calcium conditions. Time zero corresponds to the moment when germination began. R is the ratio of the calcium content of the hypocotyl of the experimental seedlings relative to that of control seedlings maintained in the standard growth solution ([Ca2+] = 2.33 mM). Seedlings were sampled for the calcium measurements on the 5th day at 17.00 hours and then 3 times a day (at 09.00, 13.00 and 17.00 hours). a Seedlings, initially grown in the standard growth solution containing a concentration of 2.33 mM calcium (Ca × 1), were transferred on the beginning of the 5th day to the calcium-depleted solution (Ca × ε); then, after 3 days, they were transferred back to the standard solution (Ca × 1). b Seedlings, initially grown in the standard growth solution (Ca × 1), were transferred on the beginning of the 5th day to an enriched solution containing 6.99 mM calcium (Ca × 3), then, after 3 days, to the calcium-deprived solution (Ca × ε) and then, again after 3 days, back to the standard solution (Ca × 1)

4.2 Evidence for Signal Memorization

When the manipulation stimulus was immediately followed by transient calcium depletion, the seedlings produced a significant number of meristems in their hypocotyl during the 2–3 following weeks (curve (a) Fig. 2). By contrast, seedlings subjected to only the calcium depletion (curve (c)) or to only the manipulation stimulus (curve (d)) produced practically no meristems. When there was a delay, δt, of 8 days between the manipulation stimulus and the beginning of the transient calcium depletion, as expected, meristems were not produced during these 8 days, but meristems were produced as soon as calcium was depleted (curve (b) Fig. 2). This data confirms (i) that the manipulation stimulus caused the storage (i.e. the memorization) of a “meristem-production” instruction within the seedlings and that calcium depletion permitted the stored instruction to be recalled and take effect by the actual production of meristems (Verdus et al. 1996, 1997; Ripoll et al. 2009). In Fig. 2, it is also apparent that the initial slope of curve (b) was steeper than that of curve (a), a result confirmed in five similar experiments (not shown). In other words, the production of meristems was more rapid when there was a delay between the manipulation stimulus and the transient calcium deletion than when calcium shortage immediately followed the stimulus. This suggests that information storage might be a relatively slow process that must be completed before the meristem-production instruction can be recalled effectively. However that may be, within each given experiment there was no systematic difference in the N values whether δt = 0, 4 or 8 days; this means that there was no appreciable loss of information during the 4 or 8 days when the meristem-production signal was stored but still not recalled.
Fig. 2

Time-course of the number, n, of meristems produced by flax seedlings when calcium depletion was delayed. The seedlings were subjected to the manipulation stimulus (down arrow) on the beginning of the 4th day, followed by a 2-day calcium depletion ([Ca2+] = ε) beginning either immediately after the manipulation stimulus (curve a, experimental points filled diamond) or 8 days later (curve b, experimental points filled square). Apart from the period of calcium depletion, the seedlings were left in the standard growth solution ([Ca2+] = 2.33 mM). Symbol t is the age (days) of the seedlings. The mean values, N (calculated according to eq. (1)), were equal to 6.0 and 5.3 (curves a and b, respectively). Curves c and d are controls in which the seedlings have been subjected to either only the calcium depletion but not the manipulation stimulus (curve c, experimental points filled triangle) or only the manipulation stimulus but not the calcium depletion (curve d, experimental points filled circle)

4.3 The Calcium-Depletion Step

Table 1 contains results typical of experiments in which flax seedlings, previously subjected to the manipulation stimulus, were exposed to a 3-day calcium depletion which was either severe ([Ca2+] = ε) or partial ([Ca2+] ranging from 0.02 to 1.75 mM). The N values were above 10 when the calcium concentration was in the range of 0–0.23 mM during calcium depletion, whereas they were close to the control values (i.e. not much above zero) when the calcium concentration was in the range of 0.58–1.75. Although in some experiments, this all-or-nothing result was less striking (data not shown), it may be concluded that calcium depletion must go below a threshold to switch on the RCL function and permit the stored instruction to lead to meristem production.
Table 1

Effect, on the seedling N value, of the calcium concentration in the growth solution during calcium depletion

Seedlings

Manipulation stimulus

Ca concentration (mM) during depletion

N values

Experimental

Yes

ε

12.1

 

Yes

0.02

11.3

 

Yes

0.23

10.3

 

Yes

0.58

1.3

 

Yes

1.17

0.9

 

Yes

1.75

0.4

Controls

Yes

2.33 (no Ca depletion)

0.1

 

No

ε

0.7

Flax seedlings were subjected to the manipulation stimulus on the beginning of the 4th day and to a 3-day calcium-depletion starting on the beginning of the 6th day. Before and after the period of calcium depletion, the seedlings were left in the standard growth solution ([Ca2+] = 2.33 mM). The severity of the calcium depletion varied ([Ca2+] ranging from ε to 1.75 mM [third column of the Table]). In the controls, seedlings were subjected either to only the manipulation stimulus ([Ca2+] kept equal to 2.33 mM) or to only a severe depletion of calcium ([Ca2+] = ε) in the absence of any manipulation stimulus

In another series of experiments, flax seedlings were subjected to the manipulation stimulus on the beginning of the 4th day, followed either by one transient depletion of calcium ([Ca2+] = ε) lasting 2–4 days, or by two successive 2-day calcium depletions. In Fig. 3, the curves (a) and (b) and the calculated N values show that the effects of a 2-day and a 4-day calcium-depletion are exactly the same. Together with previous results (Verdus et al. 1997), it may be concluded that calcium-depletion periods ranging from 0.5 to 4 days are equally efficient in the recall of stored meristem-production instruction. When the seedlings were subjected to a 2-day calcium depletion followed, 8 days later, by another 2-day calcium depletion (Fig. 3, curve (c)), as can be expected the beginning of the curve was identical to that of curves (a) and (b); however, after the second transient calcium-depletion, the seedlings started to produce a new round of meristems, which resulted in an increased N value (8.4 instead of 6) and a final number of meristems per plant almost double that with a single 2-day or 4-day calcium depletion. Similar results were obtained when these experiments were repeated with a delay of 10 days instead of 8 days between two 2-day calcium-depletion periods (data not shown). This means that the seedlings were able to recall the stored meristem-production instruction twice. With a single 6-day calcium-depletion, the representative curve (not shown) was intermediate between curves (a–b) and (c): this might correspond to the stored meristem-production instruction being recalled twice, that is, at the beginning and end of the 6-day calcium-depletion.
Fig. 3

Time-course of the number, n, of meristems produced by flax seedlings, depending on the number and duration of the periods of calcium depletion. The seedlings were subjected to the manipulation stimulus (down arrow) on the beginning of the 4th day, immediately followed by a transient calcium depletion ([Ca2+] = ε). Apart from the period of calcium depletion, the seedlings were left in the standard growth solution ([Ca2+] = 2.33 mM). The duration of the period of calcium depletion was equal to 2 days (curve a, experimental points filled triangle) or 4 days (curve b, experimental points filled square). In an additional experiment, after being treated exactly as in a [manipulation stimulus on the 4th day followed by 2-day calcium depletion], the seedlings were subjected to a second 2-day calcium depletion ([Ca2+] = ε) starting on the beginning of the 12th day (curve c, experimental points filled triangle). The N values calculated from curves a to c were a 6.0, b 6.0 and c 8.4

4.4 Applying the Manipulation Stimulus Before, During or After Calcium Depletion

The N values obtained in six independent experimental series in which calcium depletion occurred before, during or after the manipulation stimulus appear in Table 2. There are no large differences in the N values obtained under these different experimental conditions. This is a quantitative argument in favour of our previous assumption (Verdus et al. 1997) that the STO and RCL functions are independent of one another and that the effect of the RCL function is the same whether this function is switched on before or after the STO function. However, when averaging the N values obtained in the different experiments (last line in Table 2), the lowest N value was obtained when the manipulation stimulus occurred after 2 days of calcium depletion, i.e. at a moment when the Ca-concentration was especially low in the nutrient solution and in the seedling hypocotyls (Sect. 4.1). This low N value is probably not due to an effect on the RCL function but rather to an effect on the STO function since it is known that information storage is progressive (Sect. 4.2; Verdus et al. (1997)) and requires the presence of a sufficient amount of calcium (Verdus et al. 2007).
Table 2

Seedling N values according to whether the manipulation stimulus was applied before, during or after calcium depletion

Experi-mental series

Duration of Ca × ε (days)

N values in the experimental conditions

(a)

(b)

(c)

(d)

(e)

(f)

(1)

2

7.9

4.7

(2)

2

7.8

4.8

12.7

(3)

2

7.8

9.6

11.0

(4)

3

5.2

4.6

10.9

8.0

(5)

3

6.6

7.2

6.8

4.0

5.3

(6)

2

6.5

4.8

2.8

4.9

3.9

3.8

Average

7.0

6.0

4.7

6.3

6.4

8.2

Six different experimental series, numbered (1–6) (first column), were carried out. A period of calcium depletion ([Ca2+] = ε, symbolically written Ca × ε) began on the beginning of the 4th day and lasted either 2 or 3 days (second column). Apart from the period of calcium depletion, the seedlings were left in the standard growth solution ([Ca2+] = 2.33 mM). A manipulation stimulus occurred (a) immediately before the beginning of Ca × ε, (b) after 1 day in Ca × ε, (c) after 2 days in Ca × ε, (d) a few hours after returning to the standard calcium concentration, (e) 1 day after returning to the standard calcium concentration, (f) 2 days after returning to the standard calcium concentration. The average of the N values obtained in the different experimental series appears in the last line of the Table

4.5 Combining Transient Calcium Depletions with Transient Calcium Excesses

The results of a set of experiments combining transient calcium-depletions with transient calcium-excesses appear in Table 3. In the experimental series (1), when a period of calcium excess ([Ca2+] = 6.99 mM) immediately preceded calcium depletion ([Ca2+] = ε), the N value was low when the calcium depletion period was short (experiments (a) and (b)) and increased up to the N value of the positive control β (seedlings not subjected to calcium excess) when the duration of the calcium-depletion period was in the range of 6–9 days (experiments (c) and (d)). The reason is probably that, after a period of calcium excess, it takes a long time for calcium depletion to lower the calcium content of the seedling tissues enough to switch on the RCL function (Sect. 4.1; Fig. 1). In the experimental series (2), the N value for a 5-day period of excess preceding calcium depletion (experiment (e)) was appreciably smaller than that for a 2-day period of excess (experiment (f)). Again, the reason is likely to lie in the time needed to change the tissue calcium-level in the presence of calcium excesses or depletions. In the experimental series (3), the seedlings, which were stimulated either immediately before being subjected to calcium excess (experiment (g)) or after 2 days of calcium excess (experiment (h)), underwent exactly the same sequence of calcium concentrations; therefore, the large difference observed in the N values of experiments (g) and (h) (3.0 and 6.9, respectively) cannot be due to an effect on function RCL. The likely interpretation is that a high Ca-concentration at the moment when the seedlings were stimulated had a positive effect on the storage of meristem-production information (which is consistent with the observation made at the end of Sect. 4.4). In brief, high Ca-concentrations have a positive effect on the STO function (thus agreeing with our previous results (Verdus et al. 2007)), while they have a negative effect on the RCL function, and what is finally observed at the level of the N values is the result of these two antagonistic effects.
Table 3

Combination of transient calcium depletions with transient calcium excesses

Experimental series

Experimental conditions

N values

(1)

(a)

3.9

 

(b)

5.0

 

(c)

7.7

 

(d)

6.6

 

(α) [negative control]

1.4

 

(β) [positive control]

6.7

(2)

(e)

1.1

 

(f)

4.8

 

(γ) negative control

0.1

(3)

(g)

3.0

 

(h)

6.9

In series (1), experiments (a–d), the seedlings were subjected to a 3-day calcium-excess ([Ca2+] = 6.99 mM) starting at the beginning of the 4th day and immediately followed by a period of calcium depletion ([Ca2+] = ε) lasting (a) 1 day, (b) 3 days, (c) 6 days and (d) 9 days. Before and after the periods of calcium excess and depletion, the seedlings were left in the standard growth solution ([Ca2+] = 2.33 mM). Manipulation stimulus occurred on the beginning of the 6th day, i.e. during the period of calcium excess. With the negative control (α), no manipulation stimulus occurred but the experimental conditions were otherwise identical to those in experiment (d). With the positive control (β) the seedlings were not subjected to calcium excess; they were subjected to manipulation stimulus on the beginning of the 6th day and then to a 3-day calcium depletion ([Ca2+] = ε) starting at the beginning of the 7th day; before and after the period of calcium depletion, the seedlings were grown in the standard growth solution ([Ca2+] = 2.33 mM)

In series (2), the seedlings were (e) grown under conditions of calcium excess ([Ca2+] = 6.99 mM) from germination to the beginning of the 6th day, subjected to a 2-day period of calcium depletion ([Ca2+] = ε) and then grown in the standard growth solution ([Ca2+] = 2.33 mM) until the end of the experiment, or they were (f) grown in the standard growth solution from germination to the beginning of the 4th day and then treated exactly as in (e), i.e. experiencing a period of only 2 days under conditions of calcium excess (from the beginning of the 4th day to the beginning of the 6th day). The manipulation stimulus was given (e) at the beginning of the 5th day (i.e. just before the 2-day period of calcium depletion and after 4 days under calcium-excess) or (f) at the beginning of the 4th day (i.e. just before the 2-day period of calcium excess that was immediately followed by the 2-day period of calcium depletion). With the negative control (γ), no manipulation stimulus was given but the experimental conditions were otherwise identical to those in experiment (e)

In series (3), at the beginning of the 4th day the seedlings were all subjected to a 3-day calcium-excess ([Ca2+] = 6.99 mM) immediately followed by 3-day calcium depletion ([Ca2+] = ε) and, before the calcium excess and after the calcium depletion, they were left in the standard growth solution ([Ca2+] = 2.33 mM). The manipulation stimulus was given (g) immediately before the period of calcium excess or (h) at the beginning of the 6th day (i.e. after 2 days under calcium excess conditions)

In Fig. 4, the curves (a) and (b) describe the behaviour of seedlings that have been subjected to manipulation stimulus at the beginning of the 4th day, immediately followed by 2-day calcium depletion ([Ca2+] = ε) and then, after 2 days (i.e. at the beginning of the 8th day) to a 2-day (curve (b)) or 4-day (curve (a)) calcium excess ([Ca2+] = 6.99 mM), while they were left on the standard growth solution ([Ca2+] = 2.33 mM) outside the periods under depleted or excessive calcium conditions. As could be expected, after manipulation stimulus and calcium depletion, the seedlings began to produce meristems actively; but meristem production was severely reduced (curve (b)) or totally blocked (curve (a)) as a consequence of the application of the calcium-excess treatment. However, when a second 2-day calcium-depletion treatment was applied from the beginning of the 14th day (curve (c)), an active production of meristems was rapidly restored. The experiments were repeated with slight differences in the time lapses between the periods of calcium depletion and excess and similar results were obtained (data not shown). In summary, these experiments show that the RCL function, which was switched on by the calcium depletion, was switched off by the calcium-excess treatment, but could be switched on again as a consequence of a second calcium-depletion. Again (cf. the second paragraph of Sect. 4.3), this is an indication that the seedlings are able to recall several times (here twice) stored meristem-production information.
Fig. 4

Time-course of the number, n, of meristems produced by flax seedlings subjected to different combinations of calcium depletions and excesses. The seedlings were all subjected to the manipulation stimulus (down arrow) on the beginning of the 4th day, immediately followed by 2-day calcium depletion ([Ca2+] = ε). They were subjected to 2-day (curves b, experimental points filled square and c), experimental points filled triangle) or 4-day (curve a, experimental points filled triangle) calcium excess ([Ca2+] = 6.99 mM) starting on the beginning of the 8th day. The seedlings in curve (c) were subjected to a second period of calcium depletion ([Ca2+] = ε) starting on the beginning of the 14th day. Before, during and after the periods of calcium depletion or excess, the seedlings were left in the standard nutrient solution ([Ca2+] = 2.33 mM)

4.6 External Calcium-Concentration During Meristem Production

In the three series of experiments described in Table 4, the seedlings were subjected to the manipulation stimulus followed by calcium depletion, i.e. to conditions switching on both the STO and RCL functions, and then the effect of different calcium concentrations (in the range of 0.02–5.83 mM) was studied during the period of meristem production. In the three series, both low and high calcium concentrations were unfavourable to meristem production, and the highest N values were obtained with calcium-concentrations in the range of 1.75–3.73 mM, i.e. close to the calcium-concentration (2.33 mM) in our standard nutrient solution. This is an indication that, besides the roles of calcium in the action of functions STO and RCL on meristem induction, an optimal production of meristems is obtained with a intermediate concentration of calcium in the nutrient solution.
Table 4

Effect of the Ca-concentration during the production of the meristems

Experimental series

N values for Ca-concentrations (mM) during the meristem-production phase

0.02

0.23

0.58

1.17

1.75

2.33

2.91

3.03

3.50

3.73

4.08

4.66

5.83

(1)

5.8

9.7

6.1

3.5

3.8

(2)

1.1

2.6

4.2

5.2

3.7

(3)

2.6

3.1

2.9

4.4

4.6

4.9

2.8

2.1

1.4

1.6

Average

1.1

2.6

2.9

3.6

4.8

4.7

4.9

9.7

2.8

6.1

2.1

2.5

2.7

The seedlings were grown from germination in the standard nutrient solution ([Ca2+] = 2.33 mM); they were subjected to the manipulation stimulus and, some time later, to calcium depletion ([Ca2+] = ε); after that, they were left in nutrient solutions containing various Ca-concentrations. In the experimental series (1), the manipulation stimulus occurred on the beginning of the 5th day, a 2-day calcium depletion ([Ca2+] = ε) was applied starting on the beginning of the 8th day and the calcium concentration after the depletion ranged from 2.33 to 5.83 mM. In series (2), the manipulation stimulus occurred on the beginning of the 4th day, a 2-day calcium depletion ([Ca2+] = ε) was applied starting on the beginning of the 7th day and the calcium concentration after the depletion ranged from 0.02 to 2.33 mM. In series (3), the manipulation stimulus occurred on the beginning of the 4th day, a 3-day calcium depletion ([Ca2+] = ε) was applied starting on the beginning of the 8th day and the calcium concentration after the depletion ranged from 0.23 to 5.83 mM. The average of the N values obtained in the different experimental series appears in the last line of the Table

5 Discussion and Elaboration of a Conceptual Model

The data presented here agree with our previous results (Verdus et al. 1997) and are consistent with (i) a manipulation stimulus switching on a STO function that stores a meristem-production information in flax seedlings, (ii) a transient depletion of calcium switching on a RCL function that allows the stored information to be recalled and to take effect via the production of epidermal meristems in the hypocotyl and (iii) the independence of the STO and RCL functions from one another (Sect. 4.4).

The sensing of the stimulus and the storage of the corresponding meristem-production instruction was favoured by the presence of sufficiently high calcium concentrations in the nutrient solution and in the seedling tissues (Sects. 4.4, 4.5). Other authors have suggested that the sensing of an abiotic stimulus may involve external and internal calcium stores (Cessna et al. 1998). Our results reveal no appreciable loss of the stored information during at least 8 days (Sect. 4.2). They also reveal that the storage process may involve relatively slow steps (Sects. 4.2, 4.4).

Only severe calcium depletion had the potential to switch on the RCL function since depletions equal to or above 0.58 mM were ineffective (first paragraph in Sect. 4.3). Switching on the RCL function was dependent on the calcium concentration in the seedling tissues and hence only indirectly dependent on the calcium concentration in the external medium. For instance, to switch on the RCL function, seedlings previously grown in a solution with an excess of calcium required a much longer period of calcium depletion than seedlings grown in the standard nutrient solution as a consequence of the low rate of decrease of the Ca-concentration in tissues during calcium depletion (compare the calcium and meristem data in Sects. 4.1, 4.5). It is likely that it is the calcium ion that is solely important in this process, because the levels of the other ions (Na+, K+ and Mg2+) in the tissues were barely altered by the depletion treatment (Sect. 4.1).

When the first calcium depletion was followed some time later by a second calcium depletion (and independent of whether there was an excess of calcium in-between), a second round of meristem production occurred (second paragraph in Sects. 4.3, 4.5). This means that stored information can be recalled twice. A recall of information twice was also observed in experiments involving a different plant, a different stimulus and a different morphogenetic response (Desbiez et al. 1991; Thellier et al. 2000). Taken together, these data suggest that stored morphogenetic information can, as in animals (Demongeot et al. 2000 and 2006), be repeatedly solicited.

After subjecting the seedlings to a manipulation stimulus and a calcium depletion (whatever the order of these two operations and the length of time between them), the STO and RCL functions were both switched on and the production of meristems was initiated. However, the actual emergence of the meristems depended on the calcium conditions during the production period: external Ca-concentrations in the range of 1.75–3.73 mM were optimal whilst high and low Ca-concentrations were both unfavourable (Sect. 4.6). This shows that the calcium-concentration of our standard nutrient solution (2.33 mM) was in the middle of the optimal range of concentrations and thus suited to the problem under study. Taking the ensemble of the results presented here, we propose that the calcium concentration in the external medium and/or the seedling tissues has to be high enough for the storage of the meristem-production instruction, low enough during calcium depletion for the recall of the stored instruction, and neither too high nor too low for the meristems to develop once initiated. Calcium concentration thus plays several roles in the functioning of the STO/RCL system.

In previous experiments with Bidens (Desbiez et al. 1986, 1991), it appeared that circadian and non-circadian rhythms interfered with the action of the STO and RCL functions; this might mean that a physiological clock is involved in plant memorization processes. Changes in gene expression and post-translational modifications are usually considered to play the major role in plant sensitivity to environmental stimuli and in the operation of the STO and RCL functions (Davies 1987; Braam and Davis 1990; Henry-Vian et al. 1995a, b; Reyes et al. 2002; Goh et al. 2003; Tafforeau et al. 2006; Vian et al. 2006; Chinnusamy and Zhu 2009; Tran and Mochida 2010). Magnetite and phytoferritin may also play a role (Stormer and Wielgolaski 2010). However, given the number of different types of signal memorization that may exist in plants, it is difficult to understand how the complicated network of processes this would require could be entirely controlled by localized mechanisms based on multiple dedicated sets of proteins or other molecules acting independently of one another. An alternative or complementary possibility is that a delocalized mechanism is responsible (Ripoll et al. 2004, 2009). Such a mechanism might be based on ion condensation (Manning 1969, 1996; Oosawa 1971) which might help explain the multiple and subtle roles of calcium (Ripoll et al. 2004).

All these features can be brought together in a conceptual model of the storage and recall of “meristem production” information in flax seedlings (Fig. 5). This model is still at an early stage and may be incomplete or even erroneous in part. For instance, the scenario could be inversed such that the recall gene encodes a protein that is phosphorylated by the product of the DNA memory. However, the model is already consistent with many experimental observations: (i) the time-course of meristem production increases to a plateau because the amount of modified memory product responsible for meristem number increases with time as both the memory product and its recall kinase are synthesized and because the return to a normal calcium concentration then stimulates phosphatases that inactivate the memory protein and stimulates proteases that degrade the recall kinase; (ii) stored information can be repeatedly recalled because every time the calcium level is lowered there is chromatin decompaction and activation of “recall genes” which leads to a new production of meristems, (iii) the second curve is steeper than the first curve in the case when memorization occurs (Fig. 2 curves (b) and (c), respectively) because there has been time for synthesis of the memory product before the recall kinase is made, so that there is plenty of the memory product to be phosphorylated as soon as the kinase is made; and (iv) meristem induction and/or production is blocked by calcium excess because high calcium levels again stimulate a phosphatase that inactivates the memory product or stimulate a protease that degrades the recall kinase. This conceptual model should also serve as the basis for more sophisticated models to guide further experimental investigations of memory processes in plants.
Fig. 5

Conceptual model of the storage and recall of “meristem production” information in flax seedlings. On the perception of an abiotic stimulus (e.g. the manipulation stimulus), a memory is encoded by chromatin decompaction (DNA memory) possibly via the action of enzymes such as kinases, methylases or acetylases (information storage). Expression of the genes affected by the chromatin decompaction gives a “memory product”; but this product has no effect on meristem production until the recall system is activated. The recall system is under the control of calcium, possibly via calcium condensation on the chromatin keeping it compact and hence inactivating the system. High calcium thus may block the recall system, but it may also inhibit the production of the induced meristems or have both effects. Low calcium leads to chromatin decompaction and activation of the recall system via the activation of “recall genes”. The proteins (or RNA) encoded by the recall genes activate the memory product, for instance a kinase phosphorylates this memory product. In our experiments, a low calcium level was achieved by imposing a transient depletion of external calcium. Under natural conditions, low calcium may result from the combination of circadian and seasonal rhythms modifying cyclically the tissue concentrations of inorganic ions in plants. In addition to signal memorization, some signals may take effect immediately (immediate responses)

References

  1. Braam J, Davis RW (1990) Rain, wind and touch-induced expression of calmodulin and calmodulin-related genes in Arabidopsis. Cell 60:357–364CrossRefGoogle Scholar
  2. Cessna SG, Chandra S, Low PS (1998) Hypo-osmotic shock of tobacco cells stimulates Ca2+ fluxes deriving first from external and then internal Ca2+ stores. J Biol Chem 42:27286–27291CrossRefGoogle Scholar
  3. Chinnusamy V, Zhu JK (2009) Epigenetic regulation of stress responses in plants. Curr Opin Plant Biol 12:133–139CrossRefGoogle Scholar
  4. Davies E (1987) Plant responses to wounding. In: Davies DD (ed) The biochemistry of plants. Academic Press, New York, pp 243–264Google Scholar
  5. Demongeot J, Kaufman M, Thomas R (2000) Positive feedback circuits and memory. CR Acad Sci Paris 323:69–79Google Scholar
  6. Demongeot J, Thellier M, Thomas R (2006) Storage and recall of environmental signals in a plant: modelling by use of a differential (continuous) formulation. CR Biologies 329:971–978CrossRefGoogle Scholar
  7. Desbiez MO, Champagnat P, Thellier M (1986) Mécanismes de “mise en mémoire” et de “rappel de mémoire” de messages morphogènes chez Bidens pilosus L. CR Acad Sci Paris (série III) 302:573–578Google Scholar
  8. Desbiez MO, Gaspar T, Crouzillat D, Frachisse JM, Thellier M (1987) Effect of cotyledonary prickings on growth, ethylene metabolism and peroxidase activity in Bidens pilosus. Plant Physiol Biochem 25:137–143Google Scholar
  9. Desbiez MO, Tort M, Thellier M (1991) Control of a symmetry-breaking process in the course of the morphogenesis of plantlets of Bidens pilosa L. Planta 184:397–402CrossRefGoogle Scholar
  10. Goh CH, Nam HG, Park YS (2003) Stress memory in plants: a negative regulation of stomatal response and transient induction of rd22 gene to light in abscisic acid-entrained Arabidopsis plants. Plant J 36:240–255CrossRefGoogle Scholar
  11. Henry-Vian C, Vian A, Dietrich A, Ledoigt G, Desbiez MO (1995a) Change in the polysomal mRNA population upon wound signal expression or storage in Bidens pilosa. Plant Physiol 33:337–344Google Scholar
  12. Henry-Vian C, Vian A, Davies E, Ledoigt G, Desbiez MO (1995b) Wounding regulates polysomal incorporation of hsp70 and tch1 transcripts during signal storage and retrieval. Physiol Plant 95:387–392CrossRefGoogle Scholar
  13. Homes JR, Ansiaux G, Van Schoor GG (1953) L’aquiculture, 2nd edn. Ministère des Colonies de Belgique, Direction de l’Agriculture, BruxellesGoogle Scholar
  14. Manning GS (1969) The critical onset of counterion condensation: a survey of its experimental and theoretical basis. Ber Bunsenges Phys Chem 51:924–933Google Scholar
  15. Manning GS (1996) Limiting laws and counterion condensation in polyelectrolyte solutions. I. Colligative properties. J Chem Phys 100:902–922Google Scholar
  16. Oosawa F (1971) Polyelectrolytes. Marcel Dekker, New YorkGoogle Scholar
  17. Reyes JC, Hennig L, Gruissem W (2002) Chromatin-remodeling and memory factors. New regulators of plant development. Plant Physiol 130:1090–1101CrossRefGoogle Scholar
  18. Ripoll C, Norris V, Thellier M (2004) Ion condensation and signal transduction. BioEssays 26:549–557CrossRefGoogle Scholar
  19. Ripoll C, Le Sceller L, Verdus MC, Norris V, Tafforeau M, Thellier M (2009) Memorization of abiotic stimuli in plants: a complex role for calcium. In: Baluska F (ed) Plant-environment interactions: signaling and communication in plants. Springer, Berlin, pp 267–283Google Scholar
  20. Stormer FC, Wielgolaski FE (2010) Are magnetite and ferritin involved in plant memory? Rev Environ Sci Biotechnol 9:105–107CrossRefGoogle Scholar
  21. Tafforeau M, Verdus MC, Norris V, White G, Demarty M, Thellier M, Ripoll C (2002) SIMS study of the calcium-deprivation step related to epidermal meristem production induced in flax by cold shock or radiation from a GSM telephone. J Trace Microprobe Techn 20:611–623CrossRefGoogle Scholar
  22. Tafforeau M, Verdus MC, Norris V, White GJ, Cole M, Demarty M, Thellier M, Ripoll C (2004) Plant sensitivity to low intensity 105 GHz electromagnetic radiation. Bioelectromagnetics 25:403–407CrossRefGoogle Scholar
  23. Tafforeau M, Verdus MC, Norris V, Ripoll C, Thellier M (2006) Memory processes in the response of plants to environmental signals. Plant Signal Behav 1:9–14CrossRefGoogle Scholar
  24. Thellier M, Le Sceller L, Norris V, Verdus MC, Ripoll C (2000) Long-distance transport, storage and recall of morphogenetic information in plants: the existence of a primitive plant “memory”. CR Acad Sci Paris 323:81–91Google Scholar
  25. Tran LSP, Mochida K (2010) Identification and prediction of abiotic stress responsive transcription factors involved in abiotic stress signaling in soybean. Plant Signal Behav 3:255–257Google Scholar
  26. Trewavas A (2003) Aspects of plant intelligence. Ann Bot 92:1–20CrossRefGoogle Scholar
  27. Verdus MC, Cabin-Flaman A, Ripoll C, Thellier M (1996) Calcium-dependent storage/retrieval of environmental signals in plant development. CR Acad Sci Paris 319:779–782Google Scholar
  28. Verdus MC, Thellier M, Ripoll C (1997) Storage of environmental signals in flax: their morphogenetic effect as enabled by a temporary depletion of calcium. Plant J 12:1399–1410CrossRefGoogle Scholar
  29. Verdus MC, Le Sceller L, Norris V, Thellier M, Ripoll C (2007) Pharmacological evidence for calcium involvement in the long-term processing of abiotic stimuli in plants. Plant Signal Behav 2:212–220CrossRefGoogle Scholar
  30. Vian A, Roux D, Girard S, Bonnet P, Paladian F, Davies E, Ledoigt G (2006) Microwave irradiation affects gene expression in plants. Plant Signal Behav 1:67–70CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media B.V. 2012

Authors and Affiliations

  • Marie-Claire Verdus
    • 1
  • Camille Ripoll
    • 1
  • Vic Norris
    • 1
  • Michel Thellier
    • 1
  1. 1.Laboratoire AMMIS (Assemblages Moléculaires, Modélisation et Imagerie SIMS), CNRS (GDR DYCOEC), Faculté des Sciences et TechniquesUniversité de RouenMont-Saint-Aignan CedexFrance

Personalised recommendations