Abstract

The effect of osmotic stress (−0.35 MPa) on the cell water balance and apical growth was studied non‐invasively for maize (Zea mays L., cv. LG 11) and pearl millet (Pennisetumamericanum L., cv. MH 179) by 1H NMR microscopy in combination with water uptake measurements. Single parameter images of the water content and the transverse relaxation time (T2) were used to discriminate between the different tissues and to follow the water status of the apical region during osmotic stress. The T2 values of non‐stressed stem tissue turned out to be correlated to the cell dimensions as determined by optical microscopy. Growth was found to be strongly inhibited by mild stress in both species, whereas the water uptake was far less affected. During the experiment hardly any changes in water content or T2 in the stem region of maize were observed. In contrast, the apical tissue of pearl millet showed a decrease in T2 within 48 h of stress. This decrease in T2 is interpreted as an increase in the membrane permeability for water.

Introduction

Drought stress considerably reduces crop yields worldwide compared to the maximum attainable yields (Boyer, 1982; Turner, 1997). Therefore, plant responses to water deficits are of great interest in plant research, from the cellular up to the whole plant level.

In plants one of the most sensitive processes to water deficits is growth (Hsiao, 1973), making this an excellent parameter for studies of drought tolerance mechanisms and the differences between species. Although many groups have reported on the stress response of growth for roots (Ribaut and Pilet, 1991), leaves (Acevedo et al., 1971; Van Volkenburgh and Boyer, 1985) and stems (Nonami and Boyer, 1990a, b), monitoring the growth of the shoot apex has always been complicated by the use of invasive methods which impede the measurement of the apex response to processes such as drought stress at a physiologically relevant time scale.

Other drought stress responses on all organizational levels have been widely studied (Premachandra et al., 1989; French, 1997; Turner, 1997). However, until now little attention has been paid to the integration of drought stress responses at the cell, organ, and whole plant level. This is scarcely surprising since non‐invasive methods for studying whole plants on the organ or even cell level have only recently become available by means of Nuclear Magnetic Resonance (NMR) imaging.

During the past few decades NMR has been applied in a number of plant studies (Ratcliffe, 1994; MacFall and Van As, 1996; Chudek and Hunter, 1997; Ishida et al., 2000). Different from light microscopy, NMR images display the physical status and the spatial distribution of water within plant tissues in addition to anatomical information. Some examples include flow images, representing quantitative flow velocities of xylem and phloem water (Xia and Callaghan, 1992; Kuchenbrod et al., 1995; Kockenberger et al., 1997; Scheenen et al., 2000) and T2, diffusion and image amplitude (MacFall and Van As, 1996; Ishikawa et al., 1997; Donker et al., 1997; Van der Toorn et al., 2000).

The amplitude, or the initial NMR signal intensity, is directly related to the number of protons in the observed part of the system. Since in plant tissue the proton signal is dominated by water protons, the NMR signal amplitude is proportional to the tissue density times the tissue water content per voxel (Donker et al., 1997).

According to Brownstein and Tarr (Brownstein and Tarr, 1977, 1979), the observed T2 of water in a confined compartment can be described as a function of the bulk T2 (T2,bulk), the radii of the compartment along the x, y and z direction (Rx,y,z) and the rate of wall relaxation or surface sink strength density (H):  

1
formula
This model can, in principle, be applied to plant cells, with the vacuole as the main (largest) compartment with a relatively long bulk relaxation time. The cytoplasm and apoplast have a much shorter relaxation time and act therefore as a sink for the vacuole, whereas the tonoplast and plasmalemma membranes determine the exchange rates between these compartments. This relationship between T2 and permeability was also demonstrated in the diffusion simulations of Hills and Snaar (Hills and Snaar, 1992). The sink strength density (H) is defined as the rate of net loss of magnetization at the compartment surface, and will thus reflect the membrane permeabilities for water and the relaxation parameters of the cytoplasm. This model has been applied previously to spherical and cylindrical cell volumes (Snaar and Van As, 1992; Donker et al., 1997), and here it is expanded to an ellipsoid volume, where all three radii Rx, Ry and Rz can be different.

In this paper the spin–spin relaxation time T2 and the signal amplitude are used to monitor changes in the water balance of the plant stem, which should reflect adaptation processes between tissues during osmotic stress, such as membrane permeability changes and the redistribution of water. Maize and pearl millet are compared, of which the latter is more drought tolerant. The results of this study demonstrate that many aspects of plant responses to drought stress can be revealed non‐invasively by applying NMR imaging to shoot apical growth and water balance in maize and pearl millet as they respond to osmotic stress. The NMR results are combined with those of water uptake measurements, thus correlating the stress response of the apex and its surrounding tissues with the water intake and possible dehydration of the whole plant.

Materials and methods

Plant material

Maize (Zea mays L., cv. LG 11) and pearl millet (Pennisetum americanum L., cv. MH 179) seeds were germinated in wet sand in the dark. After 1 week the seedlings were transferred to water culture with half‐strengh Hoagland nutrient solution (Hoagland and Arnon, 1950), which was refreshed every week. All plants were grown in separate vessels, which were designed to fit in the NMR imager directly, to prevent root damage when the plants were transferred. The plants were grown in a phytotron at 25 °C and 70% relative humidity with a photoperiod of 16 h and a light intensity of 40 W m−2 (mercury vapour lamp). The plants were 4 or 6‐weeks‐old, with 6 or 9 fully emerged leaves respectively.

Experimental set‐up

Figure 1 shows the experimental set‐up for the water uptake measurements and NMR imaging. The roots of the plant were grown in a glass vessel, containing continuously aerated nutrient solution. The medium was pumped into the vessel from below while excess medium was removed at the top to maintain a constant medium volume in the vessel. The stem was fixed in the vessel to prevent plant movement during the measurements. The weight of the root medium vessel was measured using a balance (Ohaus GT 8000, Germany or Sartorius LP3200D, Germany). Osmotic stress was applied by replacing the normal root medium by a −0.35 MPa solution of polyethylene glycol (PEG 6000, 162.3 g kg−1) in nutrient medium. The shoot of the plant was housed in a climate chamber with an air temperature of 26±1 °C and a relative humidity of 50±2% with a photoperiod of 16 h and a light intensity of 70 W m−2.

Fig. 1.

Schematic representation of the experimental set‐up.

Fig. 1.

Schematic representation of the experimental set‐up.

MRI experiments

Three types of experiments were carried out:

  • (i)  Non‐stress conditions: The shoot apical region of maize was imaged using NMR microscopy under non‐stress conditions (N=2). Water uptake was recorded to monitor the recovery or adaptation of the plant to the new environment after insertion into the NMR imager.

  • (ii) T2 and cell size: The stems of maize (N=4) and pearl millet (N=5) were imaged from the shoot apex down over a length of about 20 cm under normal non‐stress conditions. The same plants were used for light microscopy in order to determine the parenchyma cell dimensions in the different internodes.

  • (iii) Stress conditions: NMR microscopy of the shoot apical region of maize (N=2) and pearl millet (N=2) was combined with water uptake measurements. The plants were followed for 6 d: the first 2 d under normal conditions, the second 2 d during osmotic stress and the third 2 d during recovery. For pearl millet the experiment was partially repeated (N=2), recording NMR imaging data under the initial, normal conditions and after 48 h of osmotic stress.

The NMR spectrometer was a 20.35 MHz imager consisting of a 0.47 T Bruker electromagnet (Karlsruhe, Germany) controlled by a SMIS console (SMIS, Guildford, England). In contrast to most NMR imagers, this machine has an open access in two directions, and plants can be measured in the normal, vertical position. A shielded gradient probe with an open cylindrical access of 45 mm was used (Doty Scientific Inc., Columbia, South Carolina, USA). T2 values were measured using a multi‐spin‐echo imaging pulse sequence with slice selection (Donker et al., 1997; Edzes et al., 1998), a repetition time (TR) of 1800 ms, a spin‐echo time (TE) of 4.3 ms and a spectral bandwidth of 50 kHz. A 128×128 matrix of complex data points was acquired per echo, and typically 48 or 64 echoes were acquired per echo train. For each image four acquisitions were averaged to improve the image quality, resulting in a total scan time of 15 min per image. For experiment (i) transverse images and for experiment (ii) longitudinal images were obtained, whereas for experiment (iii) both transverse and longitudinal images were acquired. The in‐plane resolution was 235 μm for the transverse images and 390 μm for the longitudinal images, both with a slice thickness of 3.0 mm.

The acquired NMR data sets were analysed by home‐written calculation routines in IDL (Research Systems Inc., Boulder, Colorado, USA). The multi‐echo images were fitted on a pixel‐by‐pixel base using a mono‐exponential decay function, yielding quantitative amplitude and 1/T2 images (Donker et al., 1996,1997; Edzes et al., 1998). After phase correction, the real part of the signal was used for analysis (Van der Weerd et al., 2000).

Light microscopy and cryo‐SEM microscopy

The same plants as used for MRM (ii) were also used to determine the cell dimensions. Fresh longitudinal sections of the internodes were prepared using a hand microtome and fixed on a microscope slide using Kayser's glycerol gelatine (Merck 9242, Darmstadt, Germany). The slices were several cell layers thick; average cell sizes for the individual plants were determined by focusing the microscope on a central cell layer and measuring 60 randomly selected cells in two dimensions for every internode by hand. The length and position of each section was measured to correlate the results to the NMR images.

Fresh longitudinal and transverse slices of the middle part of a 1 cm length stem internode of both plants were cut with a sledge microtome and embedded in Kayser's glycerol gelatine. Digital images (768×592 pixels) were made with a Sony 3CCD camera mounted on a light microscope

For the cryo‐scanning electron microscopy, internodes of 1 cm length of both plants were cut into longitudinal slices with a razor blade and plunged into liquid propane and stored in liquid nitrogen. The slices were fractured at −88 °C and freeze‐etched for 3 min in an Oxford CT 1500 HF cryo transfer unit and sputtercoated with 10 nm platinum. The cut surface was studied with a JEOL 6300F field emission scanning electron microscope at −194 °C at 2.5 kV.

Results

Non‐stress conditions (i)

Water uptake was measured for a 6‐week‐old maize plant after transfer to the imager (Fig. 2). During the first hours after transfer, the uptake rate was clearly reduced with respect to the stabilized uptake rate after a few days of adaptation. These measurements showed that the recovery of a plant after transfer took about 1 d. The adaptation of the plant following the transfer was also observable in the growth rate of the plants, which was almost zero during the first hours of the experiment and gradually increased during the first day (see Fig. 11).

Comparison of the uptake data after adaptation (Fig. 2) with similar data acquired on plants outside the apparatus shows no effect of the NMR measurements on water uptake (data not shown). Apical growth can not be measured non‐invasively on such a short time scale outside the NMR system, but the stabilization of the growth rate after adaptation in the NMR system and the agreement with the rates for internodial growth measured by monitoring the displacement of externally visible nodes every day (Whaley, 1961) indicate that the plants are not noticeably influenced by the experiments. Though long‐term effects of placing the plants in the apparatus seem to be minimal, inhibition of water uptake and growth occurs as a result of the transfer for about 1 d (Figs 2, 11). This indicates that the results of these kind of MRI studies on intact plants during the first day after setting up the experiment should be interpreted with great care.

Fig. 2.

Water uptake rate of a 6‐week‐old maize plant after transfer to the NMR apparatus. The black bars mark the dark periods.

Fig. 2.

Water uptake rate of a 6‐week‐old maize plant after transfer to the NMR apparatus. The black bars mark the dark periods.

Anatomy

Maize:

The size of the parenchyma cells in the inner part of a stem internode is fairly homogeneous (Fig. 3A), as can be observed in the longitudinal section as well (Fig. 3B). The vascular bundles are scattered throughout the ground tissue for this internode and consist mainly of protoxylem and primary phloem. The SEM picture clearly shows that the parenchyma cells are completely vacuolated (Fig. 3C).

Longitudinal and transverse NMR images for one of the plants are shown in Figs 4 and 5. In the longitudinal 1/T2 images (Figs 4A, 5) the apex is clearly visible (a). Below the shoot apex the youngest internodes can be seen (b), separated by the nodes, which appear as bright regions (c). Within the immature internodes, the T2 gradually increases from the bottom to the top. On either side of the apex, the leaves that surround the stem are visible (d). The young ear is situated directly above the apex (e), and can be seen as a grainy, brighter region (Fig. 5, e). The transverse images (Fig. 4B) clearly show the leaves (d), vascular bundles (f) and the onset of a cob (g). Note that the amplitude images show far less contrast, as expected because of a fairly uniform water content and tissue density for different plant tissues as expected on the basis of a difference in water content and tissue density.

Fig. 3.

Optical and scanning electron micrographs of a maize stem. (A) Transverse section of an internode of about 10 mm length. (B) Longitudinal section of the same internode. (C) Cryo‐SEM section of a single parenchyma cell: p, parenchyma; ph, phloem; x, xylem; v, vacuole; t, tonoplast membrane; pc, plasmalemma and cell wall; n, nucleus.

Fig. 3.

Optical and scanning electron micrographs of a maize stem. (A) Transverse section of an internode of about 10 mm length. (B) Longitudinal section of the same internode. (C) Cryo‐SEM section of a single parenchyma cell: p, parenchyma; ph, phloem; x, xylem; v, vacuole; t, tonoplast membrane; pc, plasmalemma and cell wall; n, nucleus.

Fig. 4.

Longitudinal and transverse amplitude and 1/T2 images acquired of a maize plant before and after 48 h of −0.35 MPa osmotic stress.

Fig. 4.

Longitudinal and transverse amplitude and 1/T2 images acquired of a maize plant before and after 48 h of −0.35 MPa osmotic stress.

Fig. 5.

Selection of 1/T2 images acquired of a maize plant before osmotic stress (t=0 to t=38), during −0.35 MPa osmotic stress (t=41 to t=90) and during recovery (t=96 to t=112). The light was on for 16 h daily (70 W m−2). After t=38 the stem position was altered in order to position the apex in the middle of the image plane. The dotted lines mark the beginning and the end of the stress period.

Fig. 5.

Selection of 1/T2 images acquired of a maize plant before osmotic stress (t=0 to t=38), during −0.35 MPa osmotic stress (t=41 to t=90) and during recovery (t=96 to t=112). The light was on for 16 h daily (70 W m−2). After t=38 the stem position was altered in order to position the apex in the middle of the image plane. The dotted lines mark the beginning and the end of the stress period.

Pearl millet:

As in maize, the parenchyma cells in the inner part of a stem internode are fairly homogeneous (Fig. 6A, B) and are completely vacuolated (Fig. 6C). Towards the edge of the stem, the cells become smaller. The vascular bundles are scattered throughout the ground tissue.

Though the millet stem has a similar nodular structure as maize, the anatomy as seen by NMR differs (Figs 7, 8). Most of the nodes consist of two sections (Fig. 8a), which have a higher T2 value than the internodes. Furthermore, the larger nodes are not as homogeneous as in maize and there appears to be some variety in T2 over the length of the nodes.

Fig. 6.

Optical and scanning electron micrographs of a millet stem. (A) Transverse section of an internode of about 10 mm length. (B) Longitudinal section of the same internode. (C) Cryo‐SEM section of a single parenchyma cell: p, parenchyma; ph, phloem; x, xylem; v, vacuole; t, tonoplast membrane; pc, plasmalemma and cell wall; n, nucleus.

Fig. 6.

Optical and scanning electron micrographs of a millet stem. (A) Transverse section of an internode of about 10 mm length. (B) Longitudinal section of the same internode. (C) Cryo‐SEM section of a single parenchyma cell: p, parenchyma; ph, phloem; x, xylem; v, vacuole; t, tonoplast membrane; pc, plasmalemma and cell wall; n, nucleus.

Fig. 7.

Longitudinal and transverse amplitude and 1/T2 images acquired of a millet plant before and after 48 h of −0.35 MPa osmotic stress.

Fig. 7.

Longitudinal and transverse amplitude and 1/T2 images acquired of a millet plant before and after 48 h of −0.35 MPa osmotic stress.

Fig. 8.

Selection of a number of longitudinal 1/T2 images acquired of a pearl millet plant before osmotic stress (t=0 to t=44), during −0.35 MPa osmotic stress (t=45 to t=92) and during recovery (t=96 to t=112). The light was on for 16 h daily (70 W m−2). After t=44 the stem position was altered in order to position the apex in the middle of the image plane. The dotted lines mark the beginning and the end of the stress period.

Fig. 8.

Selection of a number of longitudinal 1/T2 images acquired of a pearl millet plant before osmotic stress (t=0 to t=44), during −0.35 MPa osmotic stress (t=45 to t=92) and during recovery (t=96 to t=112). The light was on for 16 h daily (70 W m−2). After t=44 the stem position was altered in order to position the apex in the middle of the image plane. The dotted lines mark the beginning and the end of the stress period.

T2 and cell size (ii)

According to Equation (1), the observed relaxation rate 1/T2,obs is expected to depend on the dimensions of the vacuole. Figures 4 and 5 indeed show a large variation in T2 between the various internodes, which could be contributed to the different cell dimensions in these tissues. To verify whether this theoretical relationship also holds for intact plants, maize and pearl millet plants were imaged from the shoot apex down to the roots. In Fig. 9, the 1/T2 values of the various internodes of a number of non‐stressed plants (obtained from NMR images as shown in Fig. 4) are presented as a function of the cell size, which was determined afterwards by optical microscopy of the same plants. Clearly a linear correlation between cell dimensions and 1/T2 is demonstrated, that holds for a large range of cell sizes. The intercept of the fitted line in Fig. 9 corresponds to the T2,bulk of the vacuole, which is in this case around 2 s, as expected close to the T2 of free water (Snaar and Van As, 1992). According to Equation (1), the slope of the line corresponds to the sink strength parameter H, yielding H=2.8×10−5±0.52×10−5 m s−1 for maize and H=4.0×10−5±0.44×10−5 m s−1 for millet.

Fig. 9.

Relationship between relaxation time and cell dimensions for maize (n=4) and pearl millet (n=5). The data points are taken from different internodes in the NMR images. After the NMR measurements, the same plants were used for microscopic sections to determine the cell dimensions.

Fig. 9.

Relationship between relaxation time and cell dimensions for maize (n=4) and pearl millet (n=5). The data points are taken from different internodes in the NMR images. After the NMR measurements, the same plants were used for microscopic sections to determine the cell dimensions.

Maize and pearl millet during stress (iii)

The stem apical region of approximately 6‐week‐old maize plants was imaged to measure the growth rate of the shoot apex. 1/T2 images were obtained in both the longitudinal and transverse direction (Figs 4, 5). Simultaneously with the NMR experiments, the water uptake by the roots was monitored. Identical experiments were done with pearl millet.

Water uptake in maize:

After osmotic stress was applied, the uptake rate during the day decreased to about 75% of the normal rate (Fig. 10). The uptake kinetics also show a remarkable difference with those before and after osmotic stress. After the PEG solution had been replaced by normal root medium, the uptake rate increased, especially during the second day to 119% of the rate before stress was applied.

Fig. 10.

Water uptake rate of two maize plants, represented in black and grey, during −0.35 MPa osmotic stress. The dark periods are presented as black bars in the figure.

Fig. 10.

Water uptake rate of two maize plants, represented in black and grey, during −0.35 MPa osmotic stress. The dark periods are presented as black bars in the figure.

Growth in maize:

Elongation growth is presented in Fig. 11 by plotting the displacement of one particular internode versus time. Because the measured plants are not of exactly the same age, the node displacement is given in arbitrary units; the curves were scaled to have the same slope for 24 h before stress. Though the water uptake during PEG stress was only reduced by 25%, growth was almost completely inhibited during the light periods, while during the dark periods growth occurred at a very reduced rate (5–8% of the normal rate during darkness). During recovery the growth rate increased to 150% of the normal growth rate. These high rates remained unchanged for at least 24 h.

Fig. 11.

Relative growth of maize (⧫), represented in black and grey for two different plants, and pearl millet (▵), during −0.35 MPa osmotic stress. The dark periods are presented as black bars in the figure.

Fig. 11.

Relative growth of maize (⧫), represented in black and grey for two different plants, and pearl millet (▵), during −0.35 MPa osmotic stress. The dark periods are presented as black bars in the figure.

Amplitude and T2 in maize:

Before and after stress, the T2 values for the various internodes increased with time because the cell size increased due to growth. During osmotic stress, neither the T2 values nor the water density in the stem changed noticeably (Figs 4, 5; Table 1). The T2 values and the water density of the leaves (see amplitude images) showed a small decrease with time, caused by desiccation of the leaves.

Table 1.

Changes in amplitude and T2 of the central part of the 1 cm internode before stress, and after 24 and 48 h of −0.35 MPa osmotic stress

The values presented are the mean of nine voxels taken from the longitudinal images as shown in Figs 5 and 7.

 Maize
 

 

 
Pearl millet
 

 

 

 
Amplitude (a.u.)
 
1/T2 (s−1)
 
H (10−5 m s−1)
 
Amplitude (a.u.)
 
1/T2 (s−1)
 
H (10−5 m s−1)
 
 0 h PEG 5.72±0.28 2.8 4.29±0.13 4.0 
24 h PEG 0.99±0.03 5.57±0.24 2.7±0.09 0.98±0.02 4.93±0.19 4.7±0.18 
48 h PEG 0.97±0.04 5.43±0.18 2.7±0.08 1.02±0.03 5.32±0.28 5.1±0.26 
 Maize
 

 

 
Pearl millet
 

 

 

 
Amplitude (a.u.)
 
1/T2 (s−1)
 
H (10−5 m s−1)
 
Amplitude (a.u.)
 
1/T2 (s−1)
 
H (10−5 m s−1)
 
 0 h PEG 5.72±0.28 2.8 4.29±0.13 4.0 
24 h PEG 0.99±0.03 5.57±0.24 2.7±0.09 0.98±0.02 4.93±0.19 4.7±0.18 
48 h PEG 0.97±0.04 5.43±0.18 2.7±0.08 1.02±0.03 5.32±0.28 5.1±0.26 

Water uptake in pearl millet:

During PEG stress the rate decreased to about 86% of the normal rate, which is a less severe decline than in the case of maize (Fig. 12). No differences in uptake kinetics were observed either. During recovery, the uptake rate slowly increased to the normal rate, but no increased uptake compared to normal was found.

Fig. 12.

Water uptake rate of two millet plants, represented in black and grey, during −0.35 MPa osmotic stress. The dark periods are presented as black bars in the figure.

Fig. 12.

Water uptake rate of two millet plants, represented in black and grey, during −0.35 MPa osmotic stress. The dark periods are presented as black bars in the figure.

Growth in pearl millet:

During stress, growth was strongly inhibited, but not as much as in the maize plants (Fig. 11). During the day, the rate dropped to 2–3% of the normal value, whereas during the nights the rate was 12% of the normal rate. During recovery, the growth rate returned slowly to the normal rate in about 14 h, but no increased rates as compared to those before stress were observed, as were found for maize.

T2 in pearl millet:

Just as for maize, the T2 values for a particular internode increased before stress due to growth. During PEG stress however, the T2 decreased significantly in all nodes of the stem apical region to 77% of its initial value (Figs 7, 8; Table 1). Despite this decline during stress, the water density images neither showed a decrease in water density nor in stem diameter, indicating that the stem tissue did not dehydrate or shrink noticeably. The initial sink strength parameter H and the corresponding cell dimensions can be estimated from Fig. 9, using the T2 values from Table 1, yielding H=4.0×10−5 m s−1 and (1/Rlongitudinal+2/Rradial)=0.095 μm−1, respectively, for pearl millet. For maize these values are H=2.8×10−5 m s−1 and (1/Rlongitudinal+2/Rradial)= 0.186 μm−1. Assuming that the cell dimensions are constant since the water density and stem diameter did not change, the calculated radii can be used to estimate the corresponding increase in H during osmotic stress (Table 1).

Discussion

NMR microscopy

The experimental results clearly demonstrate that low field NMR imaging is an attractive tool to study stress responses in the different tissues of intact plants. By combining NMR imaging with water uptake measurements, the effects of drought stress can be monitored from the tissue up to the whole plant level. The results of the non‐stress experiments indicate that as long as plants are allowed to recover from transfer to the imager, the technique itself does not influence the plant response.

The NMR experiments reported here took 15 min per measurement. Though this is slow compared to growth measurements with, for example, an electronic transducer (Lu and Neumann, 1999), this time scale is sufficient to monitor the changes in relaxation times that occur during stress. When necessary, the time resolution can be improved further by decreasing the number of averages and the repetition time at the expense of the signal‐to‐noise ratio. At present, the time resolution of an imaging experiment is limited to a few minutes for relaxation measurements and to less than a minute for a single‐shot image.

The considerations that apply for the time resolution are also valid for the spatial resolution; i.e. improvement of the resolution can be achieved at the expense of the signal‐to‐noise ratio and/or temporal resolution.

In this study resolution at the cellular level was not obtained. However, because the tissue is so homogeneous in the internodes, it can be assumed that, in these tissues, the signal within one pixel originates mainly from cells of similar dimensions, justifying an interpretation of the information obtained in terms of cell water balance. This is especially true for the transverse images, where an optimal resolution is obtained in the axial direction (235×235 μm). The slice thickness is 3 mm, but since the longitudinal structure is so well preserved over the entire internode, every pixel consists mainly of the same tissue. The resolution of the longitudinal images is 390×390 μm, and here the slice thickness results in sampling a mix of tissues per pixel (e.g. parenchyma and vascular tissue). However, all conclusions drawn in this paper apply for both transverse and longitudinal images, i.e. the observed changes in NMR parameters occur in both types of measurements.

Growth and water uptake

The response of maize shoot growth to osmotic stress as observed by NMR is consistent with the inhibition of maize leaf and root elongation by water deficits as observed previously (Van Volkenburgh and Boyer, 1985; Ribaut and Pilet, 1991; Chazen et al., 1995). It has been suggested that the duration of the period of increased growth after stress is dependent on the duration of the preceding stress period, as has been observed for stress periods of 1 h and 4 h (Hsiao, 1973). This could account for the high growth rates for at least 48 h after recovery started.

The applied osmotic solute (−0.35 MPa PEG solution) causes only mild osmotic stress considering the relatively small change in water uptake and the constant water density in the shoot for both maize and millet. However, for both species the expansion of the shoot apical cells is strongly inhibited (Figs 4, 6).

T2 and amplitude

Though the ability to measure stem growth is valuable in itself, the major advantage of NMR lies in the additional information that can be obtained from the images. The signal amplitude is, in principle, equal to the amount of protons per voxel. However, the relaxation time of proteins and water in the cell wall matrix is too short to be observed by conventional MRI experiments, which implies that usually only intracellular water is observed (Van der Toorn et al., 2000). Thus, the signal amplitude corresponds to the tissue water content times the tissue density (Donker et al., 1997). The tissue water content is fairly homogeneous for the various organs, but appears to be the lowest in the leaves. The amplitude of the selected parenchyma tissue was constant during stress for both maize and pearl millet (Table 1), as well as the stem diameter, indicating that no noticeable dehydration nor cell shrinkage occurred under these conditions.

The relationship between T2 and compartment size (vacuole), based on the theory of Brownstein and Tarr, can indeed be observed in intact plants as well. Since there is a large variation in cell and vacuole size, this explains the large contrast in T2 between the different plant tissues. When the relationship between T2 and cell size is known, the obtained T2 values can be normalized on cell dimensions, thus revealing the role of the magnetization sink, which is determined by the permeability of the tonoplast and plasmalemma, and the cytoplasm. On the other hand, knowing that T2 increases with increasing cell size, allows growth due to cell elongation to be distinguished from growth by cell division.

In the case of pearl millet, the observed increase in 1/T2 in vacuolated cells during stress is related to such a change in magnetization sink, since the cell dimensions are constant. The sink strength density H could be calculated from Fig. 9 and corresponds to the transport of protons over the tonoplast, i.e. the tonoplast permeability for protons (free or in various molecules, mostly water), assuming a perfect relaxation sink outside the tonoplast membrane. In this in vivo situation, the relaxation sink outside the vacuole is clearly not infinite, since the relaxation rate in the cytoplasm is indeed much higher than in the vacuole, but not infinitely high. In practice, this means that H will not only reflect the tonoplast permeability (for both water and protons), but also the relaxation properties of the cytoplasm and the plasmalemma permeability. The increase found in the relaxation sink for the vacuole of pearl millet can therefore be explained either by an increase in the tonoplast and/or plasmalemma permeability for water, or by changes in the cytoplasm as a decreased T2 value in the cytoplasm due to the accumulation of sugars or a thinner cytoplasmatic layer. However, a change in membrane permeability is deemed to be more likely than the other explanations, because a decrease in cytoplasmatic T2 for these low‐field NMR measurements would require unrealistically large changes in solute concentrations (Clark et al., 1998). At a magnetic field strength of 2T, they measured the T2 of kiwifruits in different stages of ripening. The observed relaxation remained unaltered, though the soluble solids content and the free sugar content increased by 68% and 200%, respectively. At the lower magnetic field strength used (0.5T), the possible effect of solute concentrations will be even smaller.

A relationship between T2 and membrane permeability was found earlier in a heat stress study (Maheswari et al., 1999). During drought stress, an increase in membrane permeability, especially in the plasmalemma, of the elongating internodes could be advantageous, because a higher hydraulic conductivity of the internodal tissue would facilitate water transport from other plant tissues to these young, growing parts, as suggested in an earlier report (Van As et al., 1995). One can wonder if aquaporins play an important role in this change in membrane transport properties, as they are thought to contribute to drought tolerance by facilitating water mobilization towards critical cells and organs (Maurel, 1997; Tyerman et al., 1999). Recently mRNA expression of a tonoplast aquaporin was found to increase during osmotic stress in cauliflower cells (Barrieu et al., 1999).

If so, a combination of different NMR approaches, would be useful to unravel the role of membrane transport properties in the hydraulics of the system at a higher scale (tissues and organs). By relating the parameters T2 and amplitude, which provide information at the cellular level, with xylem and phloem flow measurements and with diffusion tensor imaging, to dissolve diffusion rates in all directions, in vivo information on water transport on different organizational levels is at hand.

In conclusion, maize and pearl millet differ significantly in response to osmotic stress. Though the growth rate and water uptake of pearl millet are less affected than those of maize, the effect of stress on the cell water balance of millet is more pronounced. For the interpretation of the relaxation times, a model was used which describes the relationship between relaxation, compartment size and the relaxation sink. In this whole plant approach, the T2 is indeed linearly related to compartment size. Since no changes in cell dimensions are observed for this mild stress, the observed decreases in T2 in this plant point to an increase in membrane permeability.

1

To whom correspondence should be addressed. Fax: +31317482725. E‐mail: Henk.Vanas@water.mf.wau.nl

The authors wish to thank WHL van Veenendaal, J Nijsse and AC van Aelst (Laboratory of Plant Cytology and Morphology) for help with the microscopical pictures.

References

Acevedo E, Hsiao TC, Henderson DW.
1971
. Immediate and subsequent growth response of maize leaves to changes in water status.
Plant Physiology
 
48
,
631
–636.
Barrieu F, Marty‐Mazars D, Thomas D, Chaumont F, Charbonnier M, Marty F.
1999
. Dessication and osmotic stress increase the abundance of mRNA of the tonoplast aquaporin BobTIP26‐1 in cauliflower cells.
Planta
 
209
,
77
–86.
Boyer JS.
1982
. Plant productivity and environment.
Science
 
218
,
443
–448.
Brownstein KR, Tarr CE.
1977
. Spin‐lattice relaxation in a system governed by diffusion.
Journal of Magnetic Resonance
 
26
,
17
–24.
Brownstein KR, Tarr CE.
1979
. Importance of classical diffusion in NMR studies of water in biological cells.
Physical Reviews A
 
19
,
2446
–2453.
Chazen O, Hartung W, Neumann PM.
1995
. The different effects of PEG 6000 and NaCl on leaf development are associated with differential inhibition of root water transport.
Plant, Cell and Environment
 
18
,
727
–735.
Chudek JA, Hunter G.
1997
. Magnetic resonance imaging of plants.
Progress In Nuclear Magnetic Resonance Spectroscopy
 
31
,
43
–62.
Clark CJ, Drummond LN, MacFall JS.
1998
. Quantitative NMR imaging of kiwifruit (Actinidia deliciosa) during growth and ripening.
Journal of the Science of Food and Agriculture
 
78
,
349
–358.
Donker HCW, Van As H, Jans AWH, Edzes HT.
1996
. NMR imaging of white button mushroom (Agaricusbisporus) at various magnetic fields.
Magnetic Resonance Imaging
 
14
,
1205
–1215.
Donker HCW, Van As H, Snijder HJ, Edzes HT.
1997
. Quantitative 1H‐NMR imaging of water in white button mushrooms (Agaricus bisporus).
Magnetic Resonance Imaging
 
15
,
113
–121.
Edzes HT, van Dusschoten D, Van As H.
1998
. Quantitative T2 imaging of plant tissues by means of multi‐echo MRI microscopy.
Magnetic Resonance Imaging
 
16
,
185
–196.
French J.
1997
. Primary responses of root and leaf elongation to water deficits in the atmosphere and soil solution.
Journal of Experimental Botany
 
48
,
985
–999.
Hills BP, Snaar JEM.
1992
. Dynamic q space microscopy of cellular tissue.
Molecular Physics
 
76
,
979
–994.
Hoagland DR, Arnon DI.
1950
. The water‐culture method for growing plants without soil.
California Agriculture Experimental Station Circular
 
347
.
Hsiao T.
1973
. Plant responses to water stress.
Annual Review of Plant Physiology
 
24
,
519
–570.
Ishida N, Koizumi M, Kano H.
2000
. The NMR microscope: a unique and promising tool for plant science.
Annals of Botany
 
86
,
259
–278.
Ishikawa M, Price WS, Ide H, Arata Y.
1997
. Visualization of freezing behaviors in leaf and flower buds of full‐moon maple by nuclear magnetic resonance microscopy.
Plant Physiology
 
115
,
1515
–1524.
Kockenberger W, Pope JM, Xia Y, Jeffrey KR, Komor E, Callaghan PT.
1997
. A non‐invasive measurement of phloem and xylem water flow in castor bean seedlings by nuclear magnetic resonance microimaging.
Planta
 
201
,
53
–63.
Kuchenbrod E, Haase A, Benkert R, Schneider H, Zimmermann U.
1995
. Quantitative NMR microscopy on intact plants.
Magnetic Resonance Imaging
 
13
,
447
–455.
Lu ZJ, Neumann PM.
1999
. Water stress inhibits hydraulic conductance and leaf growth in rice seedlings but not the transport of water via mercury‐sensitive water channels in the root.
Plant Physiology
 
120
,
143
–151.
MacFall JS, Van As H.
1996
. Magnetic resonance imaging of plants. In: Shachar‐Hill Y, Pfeffer PE, eds.
Nuclear magnetic resonance in plant biology
 . Rockville: The American Society of Plant Physiologists,
33
–75.
Maheswari M, Joshi DK, Saha R, Nagarajan S, Gambhir PN.
1999
. Transverse relaxation time of leaf water protons and membrane injury in wheat (Triticum aestivum L.) in response to high temperature.
Annals of Botany
 
84
,
741
–745.
Maurel C.
1997
. Aquaporins and water permeability of plant membranes.
Annual Review of Plant Physiology and Plant Molecular Biology
 
48
,
399
–429.
Nonami H, Boyer JS.
1990
a. Primary events regulating stem growth at low water potentials.
Plant Physiology
 
93
,
1601
–1609.
Nonami H, Boyer JS.
1990
b. Wall extensibility and cell hydraulic conductivity decrease in enlarging stem tissues at low water potentials.
Plant Physiology
 
93
,
1610
–1619.
Premachandra GS, Saneoka H, Kanaya M, Ogata S.
1989
. Responses of relative growth rate, water relations and solute accumulation to increasing water deficits in maize.
Journal of Plant Physiology
 
135
,
257
–260.
Ratcliffe RG.
1994
. In vivo NMR studies of higher plants and algae. In: Callow J, ed.
Advances in botanical research
 , Vol. 20. Oxford: Academic Press,
44
–123.
Ribaut JM, Pilet PE.
1991
. Effects of water stress on growth, osmotic potential and abscisic acid content of maize roots.
Physiologica Plantarum
 
81
,
156
–162.
Scheenen T, de Jager PA, van Dusschoten D, Van As H.
2000
. A model‐free quantification of water transport in intact plants with NMR imaging.
Journal of Experimental Botany
 
51
,
1751
–1759.
Snaar JEM, Van As H.
1992
. Probing water compartment and membrane permeability in plant cells by proton NMR relaxation measurements.
Biophysical Journal
 
63
,
1654
–1658.
Turner NC.
1997
. Further progress in crop water relations.
Advances in Agronomy
 
58
,
293
–338.
Tyerman SD, Bohnert HJ, Maurel C, Steudle E, Smith JAC.
1999
. Plant aquaporins: their molecular biology, biophysics and significance for plant water relations.
Journal of Experimental Botany
 
50
, Special Issue,
1055
–1071.
Van As H, Nagarajan S, van Dusschoten D, Vergeldt FJ.
1995
. Functional NMR imaging of osmotic stress in pearl miller and maize.
Plant Physiology
 
108
,
40
.
Van der Toorn A, Zemah H, Van As H, Bendel P, Kamenetsky R.
2000
. Morphological structure and water status in tulip bulbs from dormancy to active growth: visualization by NMR imaging.
Journal of Experimental Botany
 
51
,
1277
–1287.
Van der Weerd L, Vergeldt FJ, de Jager PA, Van As H.
2000
. Evaluation of algorithms for analysis of NMR relaxation decay curves.
Magnetic Resonance Imaging
 
18
,
1151
–1158.
van Volkenburgh E, Boyer JS.
1985
. Inhibitory effects of water deficit on maize leaf elongation.
Plant Physiology
 
77
,
190
–194.
Whaley WG.
1961
. Growth as a general process. In: Ruhland W, ed.
Handbuch der Pflanzenphysiologie
  XIV. Berlin: Springer Verlag,
71
–112.
Xia Y, Callaghan PT.
1992
. ‘One‐shot’ velocity microscopy: NMR imaging of motion using a single phase‐encoding step.
Magnetic Resonance in Medicine
 
23
,
138
–153.

Comments

0 Comments