Real-Time Water Movement in a Plant

Abstract

The next approach to research on water was to measure the small amount of water actually moving within a plant. The best method is to utilize radioisotope (RI)-labeled water and measure the radiation from outside of the plant. However, it is rather difficult to label water, since there are only limited kinds of RI for tracing water.

When utilizing ¹⁸F to trace water movement, another fundamental question to consider was the features that characterize drought-tolerant and drought-sensitive plants. It is natural to suppose that drought-tolerant plants have strong water absorption; therefore, by analyzing the water absorption mechanism of tolerant plants and by introducing this function to sensitive plants, it might be possible to make the sensitive plants more tolerant.

However, when water uptake was studied in naturally developed drought-tolerant and drought-sensitive cowpea, selected from 2000 cowpea plants grown in the field of Africa, the result was unexpected. Under normal conditions, the amount of water absorbed by the drought-tolerant strain was much lower than that absorbed by the sensitive strain, as if showing the low capability of water absorption. When a drought condition was introduced, the tolerant strain began to absorb much more water than usual, whereas the sensitive strain could not absorb as much water as before. This result provided us with an important lesson. Analyzing the mechanism of drought tolerance only by comparing the water absorption of tolerant and sensitive plants might not readily reveal the reason for drought tolerance. The features of the naturally produced plants showed us different mechanisms that might not match our expectations developed in the laboratory.

Next, we performed water measurements using ¹⁵O-labeled water, which has an extremely short half-life of 2 minutes. Here, we found another astonishing result, which was “water circulation” in the plant internode. A tremendous amount of water was always leaking from xylem cells, which had been regarded as a mere pipe to transfer water from the root to the aboveground parts. In another subsequent study, it was shown that the water flowing out from the xylem was pushing out the water already present in the stem and then returning to the xylem again to move upward. The water velocity in the internode was kept constant, and through simulation, it took less than 20 minutes to exchange the water already present in the stem with newly absorbed water.

1 RI-labeled water

To analyze the water absorption from the root to the aboveground part of the plant, in more detail, the basic questions are how water is absorbed by the roots, how water is transported upward from the roots, and how water interacts with the surrounding tissue of the xylem during transport. To study these questions, radioisotope tracer work is indispensable both for imaging and for the measurement of trace amounts of water.

What kind of RI can be used to label water? Since water molecules consist of two elements, H and O, there are limited radioactive nuclides applicable to label water, ³H and ¹⁵O. The method is to supply radioactive water to the plant and measure the radiation from the RI emitted by the plant. However, there is an important condition required to perform nondestructive measurement of the radioactive water supplied to a living plant. That is, the radiation emitted from the radioactive nuclides within the plant should be high enough to penetrate the plant tissue and can be detected from outside the plant by the counter prepared.

In the case of hydrogen, the only candidate radioactive nuclide for tracer work is tritium, ³H. However, the β-ray energy of ³H is too low (max. energy: 18.6 keV) to penetrate the plant tissue and be detected. Therefore, the β-rays from ³H cannot be measured from outside the plant tissue. Therefore, ³H cannot be used for the nondestructive real-time imaging or analysis of water in a living plant. Tritium can be employed only for destructive experiments. To analyze the ³H incorporated in the plant, the tissue must be removed from the plant and exposed to an IP to acquire the image, and the tissue should be digested by chemicals to measure the radiation amount by a liquid ion counter or other detector.

Then, the most promising candidate radioisotope to label water molecules is ¹⁵O, which is a positron emitter; however, there is always a problem of positron escape when using a positron emitter for imaging. Positron escape makes it impossible to calculate and compare the water amounts among the different tissues in the image. This problem will be described in more detail in the next section. Another problem in employing ¹⁵O is that the half-life is extremely short, only 2.04 min. Therefore, the experiment can last only approximately 20 min, and data must be calibrated according to the half-life.

There is another candidate for tracing water, ¹⁸F. One of the representative methods of producing of ¹⁸F is by employing the nuclear reaction ¹⁶O(α, pn)¹⁸F. Thus, when water is irradiated with an α (He) beam, trace amounts of ¹⁸F are produced in the water as a carrier-free nuclide. The carrier-free radioisotope is known to move with the surrounding molecule. Since the half-life of ¹⁸F is relatively long (110 min) compared to that of ¹⁵O (2.04 min), this nuclide was employed to measure water movement. Both nuclides, ¹⁵O and ¹⁸F, are positron emitters, and when the labeled water was supplied to the plant, the radiation from these radioisotopes in the plant could be detected from outside the plant. The radiation showed where the labeled water was, and the radiation counts could, with care, be converted to the water amounts actually present or moving within the plant tissue.

1.1 Positron Emitters

Before presenting the application of positron emitters, ¹⁵O and ¹⁸F, for tracing water movement, the features of positron emitters are presented briefly. Several kinds of positron emitting nuclides that can be applied for tracer work are produced by using an accelerator (Table 2.1).

Table 2.1 Production of positron-emitting nuclides

In the medical field, the utilization of positron emitters for nondestructive imaging of the human body is called positron emission tomography (PET). Various kinds of positron emitters are supplied to humans to diagnose lesions or tumors. However, these positron emitters have not been utilized well for plant research because of certain problems.

The radioactive nuclides called positron emitters emit positrons (β⁺) during the decay process. The emitted β⁺ soon combine with electrons (β⁻) and produce two identical γ-rays that were emitted in 180 degree opposite directions to each other at the same time, with identical energy, 0.511 keV. This phenomenon is called annihilation. In the medical field, positron emitters are widely used for two main reasons. One is that the useful positron emitters have relatively short half-lives so that after application, the radioactive nuclide will soon decay, keeping the radiation effect on humans low. The other reason is that the same detection system can be used for different positron emitters because the measurement is based on the same energy of the two γ-rays produced by annihilation, 0.511 keV. However, these measurement systems cannot be applied to plant studies since positrons easily escape from thin plant tissue. The details are described in the next section.

1.2 Positron Escape Phenomenon

As noted above, when a positron emitter is employed to visualize the distribution of the nuclide in an intact plant, there is a problem called “positron escape.” Since the positron emitted from the radionuclide has relatively high energy, the positron can move a considerable distance before undergoing the annihilation process to produce γ-rays. This means that the position of the positron emitting nuclide itself in the sample could be different from that of the γ-ray produced. The detection system focuses on the sample and measures the γ-rays from the sample, regardless of the positron escape phenomenon. The positron detector consists of a pair of γ-ray detectors for simultaneous counting for the two identical γ-rays emitted in opposite directions; therefore, the detector focuses on the sample, which is set between the two counters. In the case of a thin tissue, most of the positrons emitted from the positron emitting nuclide in the sample, could escape from the tissue and produce γ-rays in the air outside the sample. We cannot measure the γ-rays in the air, and as a result, the apparent amount of positron emitters in the image is very low.

Therefore, there is always a problem in the real-time imaging of the positron emitter in a plant because this positron escape phenomenon is dependent on the thickness of the plant tissues.

When the plant tissue is thin, like a leaf, a significant part of the positrons escape from the tissue before annihilation occurs, and as a result, the apparent amount of the nuclide at the leaf in the image is very small. As a result, positron images of a plant always show a clear image of the internode with a fading image of leaves. Therefore, to perform radioactivity counting of a positron emitter in the image of a plant over time, the counting area should be fixed at the same position in the image, and the counting at this site cannot be compared to that at other sites. Therefore, internode counting is preferable to counting the leaves.

The distance between the positron emitter site in the sample and the γ-ray production site causes the degradation of resolution. The positron escape ratio in water is shown in Fig. 2.1. Considering that more than 80% of the plant tissue consists of water, the distance between the positron emitting nuclide in tissue and the site of the γ-ray detected site after the annihilation process is sometimes on the order of mm.

Fig. 2.1

Positron escape ratio

The positron escape phenomenon also occurs in the human body in PET, and there is also a distance between the tumor, where the positron emitter accumulates, and the site where γ-rays are produced and detected. Therefore, the resolution of PET cannot be less than the order of mm. However, currently, with the assistance of image processing techniques, PET images can be modified to produce a smooth distribution of the nuclide in the tissue. However, the positron imaging system used for PET in humans cannot be applied to the plant samples, primarily because of the morphological features of the plant. A plant is largely a collection of thin sections and is very different from the human body, which is just a mass.

1.3 Production of RI-Labeled Water

Two kinds of radioisotope-labeled water are candidates to analyze the real-time water movement in a plant: ¹⁸F-water and ¹⁵O-water. Since the half-lives of ¹⁸F and ¹⁵O are short, 110 and 2.0 m, respectively, the nuclides have to be prepared just before usage and finish the experiment before decaying out the radioisotopes, i.e., while radioactivity can be detected. The preparation method of ¹⁸F and ¹⁵O is shown in Fig. 2.2.

Fig. 2.2

Production of RI for water tracers ¹⁸F and ¹⁵O

To produce ¹⁸F, 6 g of ice prepared in a Ti vial was placed in an aluminum container. Then, this container was irradiated with a He beam (50 MeV) for 40 min by an AVF cyclotron installed at JAEA (Japan Atomic Energy Agency) to produce ¹⁸F in water by the nuclear reaction ¹⁶O(α, pn)¹⁸F. After irradiation, the water was passed through a cation exchange resin to remove impurities, such as ⁴⁸V, produced in the Ti vial.

To obtain ¹⁵O, N₂ gas was irradiated with a 12 MeV deuteron (d) beam to produce ¹⁵O by the nuclear reaction ¹⁴N(d, n)¹⁵O, using an accelerator at QST (National Institutes for Quantum and Radiological Science and Technology, Japan). By irradiation, ¹⁵O-labeled gas was produced, which was then introduced to distilled water, where a Pt catalyst at 120 °C was used to catalyze an exchange reaction of ¹⁶O with ¹⁵O to produce ¹⁵O-labeled water.

As noted above, tritium, ³H, is another candidate to label water as a tracer. However, since the β-ray energy (18.6 keV) from ³H is too low to penetrate the plant tissue and cannot be measured from outside the plant tissue, ³H was not used for real-time imaging to analyze water in a living plant. Tritium was employed only for destructive experiments to acquire 2D images of water distribution by an imaging plate (IP) or for radioactivity counting by a β-ray counter after chemical digestion.

The following sections present the results we obtained by employing ¹⁸F⁻ and ¹⁵O-labeled water to trace real-time water movement in a living plant.

2 ¹⁸F-Water (Half-Life is 110 min): Cowpea, What Is Drought Tolerance?

2.1 System of ¹⁸F-Water Imaging

When water is irradiated with a helium beam, ¹⁸F is produced from oxygen, a constituent of water molecules, by the ⁶O(α, pn) ¹⁸F reaction, as described above. Considering the half-life of ¹⁸F (110 min), the tracer solution applied to the plant was adjusted to 10 MBq/ml, rather high radioactivity as a tracer.

The γ-rays emitted from ¹⁸F in the plant were measured by a pair of Bi₄Ge₃O₁₂ (BGO) scintillation detectors (Fig. 2.3). Each detector consisted of an array of small BGO detectors with a detection area of 2 × 2 mm. The aboveground part of the plant was fixed on a nylon mesh board with tape and set vertically between the two detectors. The detectors faced each other, and only simultaneous γ-ray counts by both detectors were recorded since the γ-rays were emitted at the same time by an annihilation phenomenon. The spatial resolution of the image obtained by this positron imaging system was estimated to be approximately 2.4 mm, which was the highest resolution obtained using the array of small Bi₄Ge₃O₁₂ scintillation detectors. The target area was 5 × 6 cm, and the real-time radioactivity of water in the target area was monitored with a computer. First, a cowpea plant was chosen to trace water movement within a plant using ¹⁸F.

Fig. 2.3

Schematic illustration of ¹⁸F-water imaging [1] ¹⁸F-water was supplied to the plant, and the γ-rays produced by positrons emitted from ¹⁸F were detected by a pair of BGO arrays

2.2 Cowpea

Cowpea (Vigna unguliculata Walp) was employed to study water absorption because it is widely grown in the semiarid regions of India and Africa for its high resistance to drought conditions. This plant is considered one of the most drought-resistant species among pulse crops. It was reported that under drought conditions, cowpea leaves were able to maintain comparatively high water potential and hence relatively high photosynthetic activity. As a mechanism of drought resistance in cowpea plants, it was suggested that the stem had a water storing function; however, such water storing tissue in the stem has not yet been identified, possibly due to technical difficulty.

2.3 Neutron Imaging of Cowpea

First, a neutron image of the cowpea plant was taken to determine the water distribution within the plant. As mentioned in Chap. 1, the neutron image shows a water-specific image. The neutron image acquired is shown in Fig. 2.4.

Fig. 2.4

Water image of a cowpea plant by neutron beam irradiation [2] The whiter part of the figure corresponds to the site with more water. From the extent of whiteness in the figure, cowpea was found to have a water-rich tissue in the primary leaf internode

As described in Chap. 1, the whiteness in the figure corresponds to the water amount. The whiter part of the figure corresponds to areas with more water. From the extent of whiteness in the figure, cowpea was found to have a water-rich tissue in the primary leaf internode, suggesting that this part of the plant functions as the potential water storage tissue.

Neutron images of the primary leaf internodes of soybean and common bean were taken, and only cowpea was found to have water-rich tissue in the primary leaf internode (Fig. 2.4). By comparing the cross section of the primary internode with that of the cotyledonous internode, it was found that parechymatous tissue in the primary leaf node was well developed for water storage, although a similar number of cells were counted at both internodes.

To determine the distribution of water in the internode more specifically, the whiteness of the image was dosimetrically scanned across the diameter of the primary leaf internode as well as the cotyledonous internode, and it was found that only in cowpea plants was the water content particularly high at any point in the primary leaf internode. When the internodes of cowpea, common bean and soybean were compared, no morphological differences were observed, but neutron radiography clearly showed the difference in water content (Fig. 2.5).

Fig. 2.5

Line profile of the internode in the water image [2]. Line profile of the internode in cowpea (a), common bean (b), and soybean (c) plants. The vertical axis shows the magnitude of whiteness, which indicates the water content. The horizontal axis shows the diametric distance across the internode. C: center; S: surface of the internode; Solid line: line profile across the internode between the primary leaf and the first trifoliate; Dashed line: internode between cotyledon and primary leaf. For the cowpea plant, the primary leaf internode contained a very high amount of water

Although a water storage function of the parenchymatous tissue was reported for plants growing in the desert (6,9), this cowpea plant imaging was the first direct observation of the water storage tissue.

2.4 ¹⁸F-Water Uptake of Cowpea

After the physiological properties of the cowpea plant were analyzed, ¹⁸F-water was supplied to the plant for only 1 min from the lower part of the stem, where the roots were cut off. A picture of the plant target is shown in Fig. 2.6. Then, distilled water was supplied instead of ¹⁸F-water. Since the target area was 5 × 6 cm, the area to image water uptake included the primary leaf internode and the first trifoliate leaf. The real-time radioactivity observed in the water accumulation image of the target area was monitored every 1 min. Figure 2.7 is an example of a water image in a cowpea plant. The lower figure is a photograph of the plant target, and the upper figure is an example of an ¹⁸F-water image based on a 1 min accumulation of counts. The successive water accumulation images of the target were integrated every minute until 60 min, and the results showed that water accumulated first in the primary leaf internode and gradually moved up to the first trifoliate leaf (Fig. 2.8).

Fig. 2.6

Example of setting the plant sample for imaging. The plant sample for imaging was set on a mesh and vertically between two detectors

Fig. 2.7

Cowpea plant and an example ¹⁸F-mediated water image [1]. Two cowpea plants were selected for imaging, including the primary internode (S) and the first trifoliate leaves (L). (a) Photograph of the imaging plants. An imaging area of the plants is indicated in square. (b) Example ¹⁸F-water image of the target. The size of the target area was 50 × 60 mm. The ¹⁸F signals at S and L were plotted for the absorption curve of the cowpea plant

Fig. 2.8

Successive images of the cowpea plant after ¹⁸F was supplied [1]. Successive images of the cowpea plant were acquired every 1 min of accumulation until 60 min after ¹⁸F-water was supplied from the lower part of the stem. The high ¹⁸F signal in the first image was due to the background γ-rays, since the shielding was incomplete only at this setting time

To understand the behavior of water uptake with time more clearly, the radioactivity counts of the two sites, the primary leaf internode and the first trifoliate leaf, were extracted from the series of the images in Fig. 2.8 and plotted until 30 min (Fig. 2.9). When ¹⁸F-water was supplied, the water absorption speed was very rapid at first, and then, after approximately 10 m, the increase in radioactivity plateaued. This tendency of ¹⁸F-water absorption was similar between the cowpea and common bean. The amount of water moved up to the primary leaf internode was approximately 2–3 times higher than that in the first trifoliate leaf in both plants. This difference in ¹⁸F counting between the internode and the leaf is caused by the positron escape phenomenon. As explained earlier, when the thickness of the tissue is approximately 0.2 mm, as in a leaf, approximately half of the positrons escape from the leaf tissue; therefore, radiation counting comes to approximately half that in thick tissue, such as an internode (Fig. 2.1). However, following the radioactivity at the same site made it possible to compare the change in counts with time at that site, since the escape ratio of positrons based on the tissue thickness did not change.

Fig. 2.9

Water absorption curves of cowpea and common bean plants before and after the drying treatment [1]. Successive water absorption curves of cowpea plants (Cp) and common bean plants (Cb) before and after the drying treatment. The counting regions of stem and leaf of the cowpea plant are shown in Fig. 2.7 as S and L, respectively. For the common bean, identical sites to those of the cowpea were selected

After 1 h of drying treatment, the ¹⁸F-water absorption in common bean was drastically changed from that in cowpea. Figure 2.9 shows the ¹⁸F-water uptake behavior of both plants before and after drying. Although the water transport behavior of the cowpea and the common bean was similar before the treatment, the water uptake activity of the common bean plant was drastically decreased after drying, whereas cowpea was shown to maintain high water uptake activity.

When leaf photosynthesis (LPS) activity before and after water depletion treatment in cowpea and common bean was compared, the LPS activity of the cowpea plant was maintained at a high level compared to that of the common bean (data not shown). The water image of the cowpea plant by neutron beam showed an extremely high amount of water in the primary leaf internode compared to the other internodes. The image suggested that in a cowpea plant, when water was depleted, the water in the storage tissue seemed to play an important role in maintaining both photosynthesis and water uptake activity, thereby maintaining tolerance of water-depleting conditions.

2.5 What Is Drought Tolerance?

Finally, an unexpected aspect of the water uptake activity of drought-tolerant cowpea is presented. After establishing the drought tolerance of cowpea compared to the common bean, the next question was whether there is any difference in the degree or features of drought tolerance among the cowpea. To study differences in drought tolerance within cowpea, experiments were performed on drought-resistant and drought-sensitive cowpea selected in Africa from approximately 2000 naturally grown cowpea plants. They were not artificially created by crossing, gene engineering or other methods but acquired drought-tolerant and drought-sensitive properties through evolution.

Before starting the ¹⁸F-water supply and the measurement, the nature of the tolerance was estimated. It was taken as natural that water absorption ability is the key factor in whether a plant is drought tolerant or drought sensitive. The drought-sensitive plant must have low-water absorption activity so that under drought conditions, it loses the ability to absorb water. Therefore, it is also natural to think that to make the drought-sensitive plant tolerant, its water absorption ability should be enhanced. The study then examines how the high water uptake activity in tolerant plants is regulated or maintained and determines which gene is responsible or would be effective to introduce. This approach is, of course, widely accepted an important strategy to pursue.

Using the naturally selected drought-tolerant and drought-sensitive cowpea, the fixed internode site was selected, and the water accumulating at this site was measured, considering the positron escape phenomenon. When ¹⁸F-water was supplied to the naturally selected plants, to our great surprise, the water-absorbing activity in the sensitive plant was much higher than that in the tolerant plant under normal conditions (Fig. 2.10). That is, the water-sensitive plant always required more water than the tolerant plant. However, the drought-tolerant cowpea absorbed less water than the drought-sensitive cowpea. However, after treatment with drought conditions, drought-sensitive plants could not absorb high amounts of water under normal conditions, as shown. However, the tolerant plant showed increased water absorption activity, and the amount of water absorbed was much higher than usual.

Fig. 2.10

Water absorption curves of drought-tolerant and drought-sensitive cowpea plants. Drought-tolerant and drought-sensitive cowpea plants were selected in Africa from approximately 2000 cowpea plants naturally grown in the field. Drought-tolerant cowpea commonly absorbed less water than sensitive cowpea, which was expected to be the reverse phenomenon. However, under the drought treatment, the amount of absorbed water greatly increased. The sensitive sample absorbed more water than the tolerant sample but could not absorb water after the drying treatment. Vertical axis: relative amount

This change in water absorption activity was unexpected, especially because it contradicted our assumptions about the mechanism of water absorption in the laboratory. Especially for the drought-tolerant plant, the mechanisms of maintaining lower water absorption than that of drought-sensitive plants under normal conditions and of triggering enhanced water absorption activity are not known. However, it seemed as if the tolerant cowpea was always using less energy, perhaps reserving energy for water absorption to prepare for the emergency of a drought.

3 ¹⁵O-Water (Half-Life Only 2 min): Water Circulation Within an Internode

As presented above, ¹⁸F-water measurements could trace water movement. However, there always remains a question of whether the behavior of F⁻ ions is exactly the same as that of water molecules. The number of ¹⁸F atoms produced in 1 g of water (3.3 × 10²²) targeted by a cyclotron was 3 × 10⁷. Therefore, because of the small ionic radius of F⁻ ions, the trace amount of ¹⁸F could be expected to move with the overwhelming amount of water molecules in a plant. However, to eliminate this uncertainty, the preferable radioactive nuclide to label water is ¹⁵O, with an extremely short half-life of 2 min. In the next step, ¹⁵O-labeled water was employed to study water transport in a soybean plant. As the first step to perform quantitative analysis of water movement with ¹⁵O-water, an imaging plate (IP) was employed to acquire successive static images with time. Then, a real-time measuring system for ¹⁵O-water movement was designed that enabled quantitative analysis. Since ¹⁵O is a positron emitter, like ¹⁸F, the positron escape phenomenon was consistently taken into account.

3.1 ¹⁵O-Water Image in the Internode

¹⁵O-water was produced by the nuclear reaction ¹⁴N(d, n)¹⁵O, as mentioned in 2-1-3. The ¹⁵O-water was prepared with a radioactivity concentration of 2 GBq/10 ml and supplied to the plant. Then, ¹⁵O-water imaging by an IP was performed. The plant was fixed on a board, and an IP was placed as close as possible to the board. Since the half-life of ¹⁵O is extremely short, only 2 min, the ¹⁵O-water exposure time was set to 1 min to acquire the image. Exchanging the IP with a new one, another 1 min exposure was performed to acquire the subsequent image. By comparing the two images obtained from the IPs, it was able to show the change in the ¹⁵O-water profile, which indicated water movement. As an example, ¹⁵O-water images in a soybean plant (Glycine max cv. Tsurunoko) are presented. Figure 2.11a is a schematic illustration of acquiring the RI image of the living plant.

Fig. 2.11

Usage of an IP to image ¹⁵O-water absorbed in a soybean plant [3]. A soybean plant was fixed on the board, and ¹⁵O-water was supplied from the root. An IP was placed on the board for 1 min to obtain a ¹⁵O-water image of the plant. (a) An IP was placed near the lower part of the aboveground part of the plant. To acquire the successive image, the IP was replaced with a new one after 10 and 20 min of the ¹⁵O-water supply from the root. (b) Internode image of the plant under different humidity and light conditions. R.H.: relative humidity; (c) uptake amount of ¹⁵O-water in the internode under different light intensities

After ¹⁵O-water was supplied from the root, the water image at the lower part of the internode was acquired, since the internode was a preferable tissue for quantitative analysis because of the lower occurrence of positron escape. The IP was placed on the board from 10 min to 11 min and from 20 min to 21 min after the ¹⁵O-water supply began. The images of ¹⁵O-water in the internode are shown in Fig. 2.11b. Since the half-life of ¹⁵O is extremely short at 2 min, the image of the internode soon disappeared. As shown in the figure, the darkness of the image across the internode could be converted to the amount of ¹⁵O-water absorbed. With changes in humidity and the light intensity used to irradiate the plant, the amount of water taken up by the plant changed drastically. When the relative amount of water taken up to the internode during 10–11 min and 20–21 min under 50% humidity and under the highest light intensity was set as 1, the amount of water taken up by the plant was clearly dependent on the light intensity (Fig. 2.11c).

3.2 Water Movement Is Different from that of Cd Ions

Another example of ¹⁵O-water imaging with an IP showed that the uptake of ions and that of water itself are different. In the case of Cd ion, the amount absorbed in a soybean plant was different under different pH conditions. However, there was no information on whether the amount of water absorbed was different from that of Cd ion. When the water amount absorbed is higher than that of Cd ion, it suggests that the plant is diluting the Cd concentration at absorption or during transport.

To study this plant activity, ¹⁰⁹Cd (0.325 kBq/ml) solution was supplied to an 8-day seedling of a soybean plant for 2 h under different pH conditions. Then, the distribution ¹⁰⁹Cd in the plant was imaged by an IP. The accumulation of ¹⁰⁹Cd in the aboveground part of the soybean plant was much higher under lower pH (4.5) culture conditions than under a higher pH (6.5), close to neutral. In contrast, when ¹⁵O-water was supplied and the ¹⁵O distribution was imaged by an IP, the amount of ¹⁵O-water absorbed under lower pH (4.5) conditions was lower than that under higher pH (6.5) conditions (Figs. 2.12 and 2.13). Because of the extremely short half-life of ¹⁵O, the profile of newly absorbed water was obtained for only a short time (5 min) after ¹⁵O-water was supplied. The opposite profiles of Cd and water in the aboveground part of the plant suggested that the heavy element Cd dissolved in water did not move together with the water flow in the plant. Under low pH conditions, the amount of ¹⁰⁹Cd in the aboveground part was much higher, whereas the amount of ¹⁵O-water transferred to the aboveground part was less than that at pH 6.5, suggesting that the concentration of Cd in the aboveground part was higher under acidic conditions, which might increase the toxicity of Cd in the plant. Another interpretation was that the water movement toward the aboveground part was suppressed and the upward Cd movement increased under acidic conditions. Although the IP image enables quantitative analysis, it is a static image; therefore, there was a limit to the information provided by the image.

Fig. 2.12

¹⁰⁹Cd absorption image of soybean plants under different pH conditions [4]. ¹⁰⁹Cd distribution 4 days after ¹⁰⁹Cd was supplied from the roots under different pH conditions

Fig. 2.13

¹⁵O-water absorption image of soybean plants under different pH conditions. ¹⁵O-water distribution after 5 min of ¹⁵O-water supply under different pH conditions. Although the half-life of ¹⁵O is extremely short, it showed the difference in amount of water in the aboveground part of the plant between pH 6.5 and pH 4.5. This water profile image was opposite to that of the ¹⁰⁹Cd distribution in Fig. 2.12

3.3 Real-Time Water Movement in a Plant

The ¹⁸F imaging method using an array of Bi₄Ge₃O₁₂ (BGO) detectors was difficult to apply to ¹⁵O-water imaging to analyze the real-time movement of ¹⁵O because of the positron escape phenomenon, which presented a serious problem in the image analysis. With that imaging method, it was not possible to calculate how much water was actually moving from one tissue site to the others in the image because of the different thicknesses among the tissues. Therefore, to trace the radioactivity change with time, the analysis site had to be fixed at one site in the image. Considering the features of positron emitters, the measuring system had to be redesigned to measure the amount of ¹⁵O-water moving in real time within a plant. When the imaging target was fixed to a small site of the plant (for example, see Fig. 2.10), it was possible to trace the change in radioactivity with time. Since ¹⁵O has an extremely short half-life and decays out rapidly, the counting efficiency of the γ-ray detector was required to be as high as possible, which means that a larger crystal scintillator was needed than that used for ¹⁸F. The detection area of the BOG detector was increased from 2 × 2 mm to 10 × 10 mm, and to maintain positional information, a small part of the internode, 1 cm in length, was selected. Large detectors were set adjacent to either side of the target to reduce the effect of positron escape from the sample on the detector. With this system, the volume of ¹⁵O-water in 1 cm of the internode could be measured quantitatively and noninvasively. After ¹⁵O-water was supplied to the plant, the radioactivity from ¹⁵O at the targeted 1 cm internode was measured by a pair of BGO detectors, and the amount of water accumulating in this tissue was calculated.

Figure 2.14 shows the sample at measurement and a schematic illustration of the radiation counting. When ¹⁵O decays, the nuclide is changed to ¹⁵N, and at the same time, a positron e⁺ is emitted. The positron soon reacts with an electron, e, and is converted to two identical γ-rays emitted 180 degrees apart, which is called annihilation. The pair of BGO detectors was used to count these γ-rays from annihilation, and the detector was set as close as possible to sandwich the stem (Fig. 2.15).

Fig. 2.14

Schematic illustration of ¹⁵O counting of an internode when ¹⁵O-water was supplied to the soybean. A pair of γ-ray detectors was adjusted to the position for the measurement, i.e., 2 cm above the cotyledon. Counting was performed when two identical γ-rays, which were produced from the annihilation by the positrons emitted from ¹⁵O, were simultaneously counted by both BGO detectors. The sample plants are approximately 25 cm in height. After the application of ¹⁵O-water, the ¹⁵O-water amount accumulated at 1 cm of the internode was counted. The half-life of ¹⁵O is 122 s

Fig. 2.15

Diagram of the measurement system for ¹⁵O-water. A long cable (approximately 4 m) was used to connect the gamma-ray detector with the coincidence circuit. The vial and detectors were shielded by lead blocks. The illustrated parts, except for the coincidence circuit, were set in the growth chamber. PMT: photomultiplier tube

3.4 Design of ¹⁵O-Water Measuring System

The system to measure ¹⁵O-water accumulation at the internode of the plant was designed as shown in Fig. 2.15. Figure 2.16 shows a schematic illustration of a BGO detector (crystal size 10 × 10 × 20 mm, Photosensor Modules: Hamamatsu Photonics, Co., Japan) and a picture of the detector placed adjacent to the plant internode. A BGO crystal was coupled to photomultipliers fixed in an aluminum frame.

Fig. 2.16

BGO detector. (Photograph of the plant target with a pair of BGO detectors. The detector consisted of a BGO crystal, a PMT, and a preamplifier. The BGO crystal was coupled to photomultipliers fixed in an aluminum frame. The dimensions are given in mm)

A pair of BGO probes was installed face to face across the internode, which was 2 cm above the cotyledon of the plant. The signals that entered the detection window were amplified by a linear amplifier (704-4B Oken, Co., Japan), and then discriminated by a Timing S.C.A. (706-2B: Oken, Co.) according to the γ-ray energy of an annihilation, 511 keV. Then, the signals were converted to counts through a coincidence circuit (Fast and Slow Coincidence 708-1B, Oken, Ratemeter S-2293B, Oken, Co.) and recorded in a computer (OptiPlex GX1: Dell. Co. Kawasaki). The timing of coincidence and the data export interval were set at 110 ns and 1 s, respectively. The background noise was recorded to <0.01 cps by the coincidence circuit and shielding made of lead blocks.

Since the radioactivity of ¹⁵O is reduced by half every 120 s, the radioactivity counting of ¹⁵O–water was able to continue for approximately 1000 s. The counts were accumulated every 30, and the detection limit was estimated to be 0.11 kBq from the least squares fitting curve of observed counts (Fig. 2.17). The counting efficiency of the γ-ray detector was calculated to be 0.120% by comparing the counts of the BGO probes with that of the γ-ray counts by a counter for the plant tissue supplied with H₂¹⁵O (200 MBq/ml⁻¹). Table 2.2 summarizes the performance of the measuring system using the BGO detector. This performance of the detector was also confirmed by preparing a phantom of the stem consisting of a silicon tube containing ¹⁵O-water gel close to the surface, mimicking the soybean stem with xylem cells (Fig. 2.18).

Fig. 2.17

Detection limit of the measurement system. The radioactivity of ¹⁵O-water was counted for 1000 sec, and the counts were accumulated every 30 s. Broken line: background level; solid line: background level + 2σ; broken curve: least-squares fitting exponential curve of the observed counts

Table 2.2 Performance of the measuring system using the BGO detector

Fig. 2.18

Phantom of the prepared stem. To calibrate the γ-ray counting to the water amount, in addition to the actual gamma-ray counting by cutting the internode, the phantom of the internode was prepared, and calibration was confirmed

Figure 2.19 shows an example of ¹⁵O-water counting in a soybean stem when ¹⁵O-water was supplied. As shown in Fig. 2.19a, the counts from the internode increase at first and then decrease rapidly because of the short half-life of ¹⁵O. Therefore, the ¹⁵O-water absorption curve was always calibrated with half-life decay, and the curve in Fig. 2.19a was converted to that in Fig. 2.19b.

Fig. 2.19

Calibration of the ¹⁵O count decay. (a) Raw counting data measured by the system. The data were obtained every 10 s; (b) ¹⁵O-water uptake curve calibrated with the half-life decay and counting efficiency. The data were calculated every 30 s

The measurement system was set in a phytotron, which maintained the conditions throughout the experiment at 27 °C and 50% humidity, and all experiments were performed during the light phase.

3.5 ¹⁵O-Water Absorption Curve

The prepared plant sample was a soybean plant 20 days after germination with second expanded trifoliate leaves. The plant height was approximately 25 cm. After eliminating the cotyledon and excising the root 8 cm below the cotyledon, the bottom part of the stem was placed in a vial. Then, ¹⁵O-water (2 GBq/10 ml) was supplied to the vessel. The vessel and the detectors were shielded by lead blocks to reduce background counts. After ¹⁵O-water was supplied to the plant, the amount of ¹⁵O-water in the 1 cm internode increased linearly at first, and then the slope gradually decreased. Since the half-life of ¹⁵O is only 2 min, measurement could be performed until approximately 1000 s. The calibrated ¹⁵O-water amount plotted in the absorption curve is shown in Fig. 2.20. Surprisingly, the amount of ¹⁵O-water measured at the 1 cm internode exceeded the xylem volume (1.9 μL) in the internode within only one minute.

Fig. 2.20

¹⁵O-water uptake curve at 1 cm stem of a soybean plant [5]. ¹⁵O-water was supplied to a soybean plant, and the amount of the water in 1 cm of the stem, which was 2 cm above the cotyledon, was measured. Since the half-life of ¹⁵O is extremely short (2 min), the absorption measurement could be perform until approximately 1000 s. Within a few minutes, the amount of ¹⁵O-water in the stem exceeded the vessel volume in the 1-cm stem (2 μL) and increased to approximately 45–55 mL, which is close to the entire volume of the targeted 1-cm stem, after 1000 s. The results indicate that a large amount of water leaked from the xylem. Each symbol represents the data of individual plant samples (N = 5)

The capacity of the xylem vessels was calculated from the measurement of their transverse sectional area under microscopy {1.9 ± 0.3 × 10⁻⁷ m² (total cross section area) × 1 × 10⁻² m (length of measuring part of the internode)}. From the linear part of the slope of the absorption curve, the increasing amount of ¹⁵O-water per 1 cm of the internode was calculated as 5.2 ± 0.5 × 10⁻² μL/s.

During the measurement, the amount of ¹⁵O-water continued to increase and occupied a volume of 40 μL after approximately 15 min, which was more than 20 times higher than the vessel capacity and close to the whole volume of the targeted stem, 1 cm in length (45–55 μL).

This ¹⁵O-water absorption curve indicated a very interesting result: a tremendous amount of water was always leaking out from the xylem vessel, which had been regarded as a mere pipe to transport water to the surrounding tissues. It was also suggested that the water was leaked out not longitudinally but horizontally. However, continuous lateral water movement must be associated with longitudinal transport, which was derived from transpiration by the plant.

3.6 Route of Water Flow Leaked from Xylem

From the measurement of the newly absorbed water movement and accumulation in the internode, it was found that a large amount of water was always leaking out from xylem vessels (Fig. 2.21). The next question was the route of the water after leaking out from the xylem. There are four possible routes for the flow of water leaked from xylem.

• 1. Flow out from the stem surface.

• 2. Flow into the phloem vessel.

• 3. Move upwards in other tissue besides the xylem vessel.

• 4. Return to the xylem vessel.

• 1. Flow out from the stem surface.

  When water moves toward the surface of the internode and evaporates there, it could provide a driving force for water to move within the internode. To determine whether the water leaked out from the xylem vessel was lost from the internode surface, the surface of the internode was covered with Vaseline, and the ¹⁵O-water uptake curve was measured. There was no change in the water absorption curve before and after covering the internode with Vaseline (Fig. 2.22). The results indicated that the large amount of water leaking out from the xylem vessel was not lost from the internode surface, which means that most of the leaked water was not moving toward the surface of the internode.

• 2. Flow into the phloem vessel.

  Since there are some reports that there is an interaction between xylem and phloem vessels, the outside of the cambium was removed, and the water uptake curve was measured. Figure 2.23 shows that there was no change in the water uptake curve after removing the phloem vessels, indicating that water leaked from the xylem vessels was not moving toward the phloem vessels.

• 3. Move upwards by other tissue besides the xylem vessel.

  Dissection of the internode of the stem showed clearly that the inside of the xylem vessel was empty, and the other tissues were filled with tissue cells. Therefore, it can easily be estimated that it might be natural for water to move upward within the xylem vessel than through other tissue where resistance significantly decreases the movement speed of the water.

  If the leaked water is moving upward through the tissue other than the xylem vessel, there must be a difference in the velocity of the water movement from that in the xylem. To determine the water velocity within the xylem, the volume of the xylem in the internode was measured. Since the inside of the xylem vessel is empty, when there is no change in the whole volume of the xylem vessel with height, it can be concluded that there is no change in the water velocity within the xylem vessel. When the internodes of the stem, 2 and 6 cm above the cotyledon, were sliced and the area of the xylem vessels was measured under a microscope, the area of the xylem vessels at different heights was almost the same, 1.91 ± 0.32 and 1.86 ± 0.31 (×10⁻⁷ m²), respectively. This result indicated that the same volume of water was constantly transferred upward within xylem vessels, regardless of the height in the internode, suggesting that the speed of upward water movement in the xylem vessels is the same throughout the internode. The next question concerns the volume of the tissue surrounding the xylem tissue containing the water. As shown in the absorption curve, the ratio of the water volume in the xylem to that in the other tissue, flowing out from the xylem, was approximately 1:20, indicating that the leaked water volume was much higher than that remaining in the xylem. Under the microscope, the area of the xylem vessel and the other tissue was measured, and it was found that the ratio of xylem vessel area to that of the other tissue was approximately 1:20, which could be converted to volume. That means that approximately 20 times more water was contained in the 20 times larger volume of the tissue other than xylem.

  Since the ratios of the amount of water and the tissue volume in xylem and the other tissue were approximately the same, the upward velocity of water movement in each tissue must be the same when the leaked water was moving through other tissues. However, as noted above, it cannot be the same because of the structure of xylem and the other tissue.

  There is another reason to discount this model, which is the water velocity in the internode shown in Fig. 2.25, where the water velocity in xylem cells remained the same (see next section). In conclusion, this possibility was ruled out as an explanation of the behavior of leaked out water, although the evidence for its rejection is indirect.

• 4. Return to the xylem vessel

  Of the four possibilities for the route of water movement after leaking out from xylem vessels, 1 to 3 were rejected, as discussed above. Therefore, No. 4 remained to explain this phenomenon. That is, most of the water leaking out from the xylem vessels must re-enter the xylem vessels. According to this scenario, the decrease in the gradient of the ¹⁵O-water absorption curve along with the height of the measuring position of the internode has to be interpreted as a decrease in the ¹⁵O concentration in xylem vessels rather than a decrease in water escape.

Fig. 2.21

Schematic illustration of four possible routes for water movement after the escape from the xylem vessels: ➀ water flows upwards through the tissue other than xylem vessels; ② water flows towards the stem surface and evaporates; ③ water flows into the phloem tubes and is transported downward; ④ water re-enters the xylem vessels

Fig. 2.22

Effect of the Vaseline treatment on the lateral water movement [5]. Water evaporation from the stem surface was inhibited by a Vaseline cover on the surface. The vertical axis indicates the volume in the targeted stem occupied by ¹⁵O-water. ¹⁵O-water was applied at t = 0. The filled circles and open squares quantify the volume of ¹⁵O-wate before and after the Vaseline application, respectively

Fig. 2.23

Water uptake curve when cambium was removed [5]. Phloem flow was knocked out by removing the tissue outside of the cambium at 3.0–3.5 cm and 1.5–2.0 cm above and below the measuring position, respectively. The vertical axis indicates the volume in the 1-cm stem occupied by ¹⁵O-water. ¹⁵O-water was applied at t = 0. The filled circles and crosses indicate the volume of ¹⁵O-water before and after removing the tissue outside of the cambium, respectively

3.7 Water Flow in the Internode

In the previous section, it was found that freshly absorbed water was circulating within the stem, leaking out from the xylem and returning to the xylem by exchanging water already present in the stem with the newly absorbed water. To determine the exchange behavior of newly absorbed water in xylem with that in the surrounding tissue, ¹⁵O-water was supplied for 5 min as a pulse, and then nonradioactive water was supplied to the soybean plant. As shown in Fig. 2.24, ¹⁵O-water linearly increased for approximately 300 s and then slowly decreased, indicating that the water from xylem was smoothly exchanged with the water already present in the internode. This ¹⁵O-water movement suggested that the ¹⁵O-water already spread out horizontally at the internode in 5 min was pushed horizontally toward the xylem by newly absorbed nonradioactive ¹⁶O-water, acting as a returning flow, and then moved upward. As a result, after its initial increase, the ¹⁵O activity gradually disappeared from the fixed measuring site of the internode.

Fig. 2.24

Exchange of ¹⁵O-water with stable water and ¹⁶O-water. Measurement of ¹⁵O at 2 cm above the cotyledon. After ¹⁵O-water was supplied for 5 min, stable water (¹⁶O-water) was supplied. After the water was changed from ¹⁵O-water to ¹⁶O-water, the ¹⁵O-water amount at the internode continued to increase for approximately 200 s and subsequently gradually decreased, which suggests that water exchange via xylem vessels smoothly occurred

Another question is about the vertical water flow in the internode. To measure the water movement speed at different heights of the internode, three pairs of BGO detectors were prepared to measure the water uptake curve. Figure 2.25 shows the ¹⁵O-water uptake curve measured at different positions of the internode, 35, 50, and 65 mm above the cotyledon (a, b, and c). From the difference in the time when the detectors first detected the ¹⁵O-water, d, e, and f, the water transfer speed in the internode could be calculated. As shown in the figure, the time needed for water to move 15 mm, that is, from a to b and from b to c, was shown to remain the same, indicating that the water movement speed in the internode remained constant (4 mm/s).

Fig. 2.25

Flow rate of ¹⁵O-water in the internode [5]. The amount of ¹⁵O-water absorption curve was measured at three positions in the internode: 35, 50, and 65 mm above the cotyledon (a, b, and c, respectively). The intersection of the curve at the x-axis after elimination of the noise shows the time when ¹⁵O-water was first detected, as indicated by d, e, and f. From the time difference between d and e and between e and f, the flow velocity of water was constant in the internode and calculated to be approximately 4 mm/s

Assume that the xylem vessels can be regarded as mere pipes with many surface pores and that water is transported through them. When water escapes from the pipe and does not return to it, the amount of water transported should decrease with height along the pipe. As shown in Fig. 2.25, the flow velocity of ¹⁵O-water through the xylem vessel remained almost constant. This result also supported that the water leaked from xylem vessels was returning to the xylem again.

Finally, another experiment was introduced, which simply measured the respiration volume of water from the plant. The evaporation of water was measured by weighing the whole plant in a phytotron, and the evaporation speed was 0.91 ± 0.13 μL/s, including leaves and internodes. As described above, the ¹⁵O-water uptake experiment showed that the speed of water leaking from the xylem vessel was 0.052 μL/s/cm (Fig. 2.20). Since the whole internode length was approximately 16.5 cm, the total leaked volume of water from the whole internode was approximately 0.86 μL/s, which was close to the amount of water lost by respiration measured by the weighing experiment, suggesting that the water volume leaking from xylem vessels corresponds to the decrease in water resulting from respiration.

3.8 Verification of Water Returning Process to Xylem Using ³H-Water

Another remaining question concerns the horizontal movement of water within the internode. To verify the process of water leaking from and returning to xylem tissue, ³H-water was used instead of ¹⁵O-water. Since the β-ray energy emitted from ³H is very low, 18.6 keV, this radiation cannot be detected from outside the plant when ³H-water is supplied to the plant. Therefore, several plants were prepared, and ³H-water was supplied for 5 s to the plants as a pulse. Then, periodically, the internode at the same position of the ¹⁵O-water measurement was cut out, and the ³H-water distribution image was acquired by an IP.

Figure 2.26 shows the image of the ³H-water distribution in the internode section after 0, 10, 20, 60, and 120 s of ³H-water supply, with the corresponding picture of the dissection image under a microscope. High accumulation of ³H-water was observed in xylem vessels as early as 5 s of application (time 0 s in Fig. 2.26). After 20 s, ³H-water was spread throughout the transection of the internode. This indicated a rapid inward movement of ³H-water, independent of transpiration. Then, the ³H-water amount decreased, and the intensity of the ³H-water image decreased greatly, indicating that the leaked water returned to the xylem again and moved upwards. Water diffusion is another candidate for water movement. When the diffusion coefficient was 2.4 × 10⁻⁹ (m²/s), the diffusion distance of ³H-water for 10 cm of the stem could be less than 2 mm after 10 min. Therefore, it was suggested that most of the ³H-water must have been transported longitudinally through the xylem, followed by horizontal diffusion at the point of the measurement.

Fig. 2.26

³H-water uptake shown in the stem sections [5]. To confirm the horizontal leak of water using ¹⁵O-water, ³H-water was similarly supplied for 5 s (t = 0). Since the β-ray from ³H is too low to detect from outside of the plant, imaging could not be nondestructively performed. The stem was harvested and sliced each time after the ³H-water supply, and the microradiograph of the sliced section was acquired by IP. (a) Image of the ³H-water distribution in the internode section, 2 cm above the cotyledon, from the end of the ³H-water supply (0 s) to 120 s. (b) Microscopic images corresponding to the upper images. ³H-water leaked from xylem was spread throughout the area of the section after 20 s and gradually returned to xylem tissue

In short, the results showed that the leaked water spread out horizontally in a short time, flushing away the water already present in the internode and then returning to the xylem vessel again moving upward within the internode.

3.9 Summary of Water Circulation Within the Internode

Using both ¹⁵O-labeled and ³H-labeled water, the water movement in the internode showed that there was an intense and ongoing lateral water exchange between the xylem vessel and the surrounding tissues along the upward pathway. Therefore, the ¹⁵O-water uptake amount decreased because of dilution of the ¹⁵O-water in xylem tissue with the nonradioactive water flushed out and returned to the xylem vessel during the ¹⁵O-water absorption process. Simulation showed that approximately half of the water already present in the stem was replaced by freshly absorbed water within approximately 20 min (data not shown).

The results of water movement in a soybean plant can be summarized as follows.

• 1. A tremendous amount of water was always horizontally leaking from xylem tissue.

• 2. Water that leaked from the xylem tissue then flushed away the water already present in the internode, re-entered the xylem tissue and moved upward, showing water circulation in the stem.

• 3. The velocity of upward water movement remained constant.

• 4. In a simulation, within 20 min, half of the water already present in the internode was estimated to exchange with the freshly absorbed water.

4 Summary and Further Discussion

Radioisotope-labeled water provided water absorption movement in a living plant. As a representative study, ¹⁸F-water uptake in a cowpea plant and ¹⁵O-water uptake in a soybean plant are presented.

In the case of a cowpea plant, a special internode was identified, and the water movement under water-deficient treatment suggested that this tissue functioned as water storage tissue to enable heat tolerance. However, it was noted that the water uptake behavior of drought-tolerant and drought-sensitive cowpea naturally produced in the field of Africa showed water-absorbing activity that could not be estimated before. The water-absorbing activity of drought-tolerant cowpea naturally produced in the field was lower than that of the drought-sensitive cowpea, but the water absorption activity of the drought-tolerant cowpea increased greatly under drought conditions, whereas the sensitive cowpea showed the opposite effect. The plant strategy for tolerating water depletion suggested that enhancing water absorption activity is not the way to achieve drought tolerance; instead, growing with limited water requirements is necessary. However, why the water absorption activity of the tolerant plant was suddenly enhanced upon exposure to low-water conditions is still not known.

Although ¹⁸F is produced as a carrier-free ion by irradiating water with a ⁴He beam, whether ¹⁸F⁻ has the same behavior as that of water itself remains uncertain. Therefore, ¹⁵O was produced and used to trace water movement in a soybean plant. Since the half-life of ¹⁵O is extremely short, 2 min, the experiment was performed before the ¹⁵O decayed out, within approximately 20 min.

Using ¹⁵O-water, it was found that a tremendous amount of water was constantly leaking out from the xylem and returning to the xylem again after flushing out the water that was already present in the xylem vessel. This circulation of water in an internode was observed for the first time. The renewal time of the water already existing in the internode was rather fast. It took approximately 20 min to replace half of the water already present in the internode with newly absorbed water, according to a simulation. Since an internode is packed with cells, and each cell has a cell membrane and organelles inside, it is totally unknown how the water could enter these different micro-organelles easily and flush out the water already present. The only known water movement channel is aquaporin. Are there networks of different kinds of aquaporins in the internode?

The xylem vessel is apt to be regarded as a mere pipe to transfer water from the root to the upper tissues. However, the vessel itself has a microscopic structure that differs from place to place. Figure 2.27 shows two cross sections of an internode of a soybean plant, only 2 mm apart. Within 2 mm of distance, some of the vessels disappear, and some new ones appear. This means that the structure of the xylem vessels forms a complicated network throughout the internode, with microscopic changes in the morphological shape and size of the vessels or connection sites with height. However, it was interesting to observe that not only the area of vessels but also the areas of the sieve tube, pith and xylem remained in the same range even when they were 4 cm apart in the internode (Fig. 2.28). That means that the same volume of structure is maintained throughout the internode, suggesting that the function and the activities within the internode remain constant, such as transferring the same volume of water.

Fig. 2.27

Horizontal section of the soybean stem under a microscope. To confirm the location and size of xylem tissues, sliced sections at different heights are shown. The locations of A and B are 2 mm apart from each other. The location and size of each xylem tissue are different. Numbers 1 and 2 3 in the same color are the same xylem tissue

Fig. 2.28

Tissue area in the section of the stem at different heights. The stem was harvested from 2 and 6 cm above the cotyledon of a soybean plant. The areas of the vessel, sieve tube, pith, and xylem were measured under a microscope. As shown in Fig. 2.27, although the shape and position of the xylem tissue varied, the area at different heights was approximately identical, which supports that the same amount of water was constantly transferring upwards

The secondary cell wall of the xylem vessel consists of a thick wall containing lignin, which becomes thicker as the plant grows. Under a microscope, it was observed that the thick secondary wall of lignin surrounds the surface of the vessel tube, like a bellows, to provide mechanical support for its function of water or nutrient transport in the internode. Figure 2.29 shows an example of the vessels in a soybean plant. In the case of the thick wall, the bellows spacing became narrower with the development of the plant, suggesting that the volume of water leakage might be controlled by this spacing. Actually, at senescence, hardly any space for this bellows was observed surrounding the vessel (data not shown).

Fig. 2.29

Xylem structure under the microscope

Since the flexible movement of the bordered pit field functions to adjust water movement, a question was raised about the water stream within the vessel: whether the water moves straight up or if there is any whirling motion in the vessel. To understand the upward flow of water within the internode, the Reynolds number in the vessel was calculated. It was a very small number, 0.002–0.003, given 20 μm as the diameter of the vessel, 1 mm/s as the water velocity and 10⁻⁶ m²/s as the kinematic viscosity, suggesting that the water flow within the vessel along the vessel wall was not creating any whirling stream.

Another question is whether there was isotope exchange between ¹⁵O-water and natural ¹⁶O-water in the vessel. To test this possibility, 10 ml of ¹⁵O-water was passed through the column packed with 1.5 g of cellulose, the water coming out from the column was collected every 300 μL, and the ¹⁵O-water concentration was measured. When isotope exchange occurs, the ¹⁵O-water concentration in the collected fraction must be decreased. The ¹⁵O-water decreased by 14% only in the first fraction, and there was hardly any decrease in the ¹⁵O-water concentration in subsequent fractions. Therefore, isotope exchange within the internode was not taken into account in our experiment.

The motive force of water leakage from vessels and the subsequent lateral movement of water remains a major question. The pressure inside the vessel is low, approximately 0.8 M Pa. In spite of this low pressure, the water flowed out from the vessel horizontally. The pith might be one of the candidates for the horizontal route. As shown by the ³H-water image after 20 s (Fig. 2.25), water moved horizontally toward the pith according to diffusion. Another question is the relation between the leaked water and that lost by transpiration, i.e., the relation between horizontal and vertical water flow.

Last, the water movement found here provided us with many further questions regarding how plants regulate the water flow within the plant, particularly what causes water leakage from xylem tissue and its return to the xylem, which seems to be an important and basic plant activity.