Journal of Experimental Botany, Vol. 55, No. 396, pp. 433-447, February 1, 2004
© 2004 Oxford University Press
Regulation of Growth, Development and Whole Organism Physiology |
Water permeability and reflection coefficient of the outer part of young rice roots are differently affected by closure of water channels (aquaporins) or blockage of apoplastic pores
Received 6 June 2003; Accepted 17 October 2003
1 Lehrstuhl Pflanzenökologie, Universität Bayreuth, Universitätsstrasse 30, D-95440 Bayreuth, Germany
2 International Rice Research Institute, DAPO 7777, Metro Manila, Philippines
* To whom correspondence should be addressed. Fax: +49 921 552564. E-mail: ernst.steudle{at}uni-bayreuth.de
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionReferences |
|---|
The relative contribution of the apoplastic and cell-to-cell paths to the overall hydraulic conductivity of the outer part of rice roots (LpOPR) was estimated using a pressure perfusion technique for 30-d-old rice plants (lowland cultivar, IR64, and upland cultivar, Azucena). The technique was based on the perfusion of aerenchyma of root segments from two different zones (20–50 mm and 50–100 mm from the root apex) with aerated nutrient solution using precise pump rates. The outer part of roots (OPR) comprised an outermost rhizodermis, an exodermis, sclerenchyma fibre cells, and the innermost unmodified cortical cell layer. No root anatomical differences were observed for the two cultivars used. Development of apoplastic barriers such as Casparian bands and suberin lamellae in the exodermis were highly variable. On average, matured apoplastic barriers were observed at around 50–70 mm from the root apex. Lignification of the exodermis was completed earlier than that of sclerenchyma cells. Radial water flow across the OPR was impeded either by partially blocking off the porous apoplast with China ink particles (diameter 50 nm) or by closing water channels (aquaporins) in cell membranes with 50 µM HgCl2. The reduction of LpOPR was relatively larger in the presence of an apoplastic blockage with ink (
30%) than in the presence of the water channel blocker (10%) suggesting a relatively larger apoplastic water flow. The reflection coefficient of the OPR (
sOPR) for mannitol significantly increased during both treatments. It was larger when pores of the apoplast were closed, but absolute values were low (overall range of sOPR=0.1–0.4), which also suggested a large contribution of the non-selective, apoplastic path to overall water flow. The strongest evidence in favour of a predominantly apoplastic water transport came from the comparison between diffusional (PdOPR, measured with heavy water, HDO) and osmotic water permeability (PfOPR) or hydraulic conductivity (LpOPR). PfOPR was larger by a factor of 600–1400 compared with PdOPR. The development of OPR along roots resulted in a decrease of PdOPR by a factor of three (segments taken at 20–50 and 50–100 mm from root apex, respectively). Heat-killing of living cells resulted in an increase of PdOPR for both immature (20–50 mm) and mature (50–100 mm) root segments by a factor of two. Even though both pathways (apoplast and cell-to-cell) contributed to the overall water flow, the findings indicate predominantly apoplastic water flow across the OPR, even in the presence of apoplastic barriers. Low diffusional water permeabilities may suggest a low rate of oxygen diffusion across the OPR from aerenchyma to the outer anaerobic soil medium (low PO2OPR). To date, there are no data on PO2OPR. Provisional data of radial oxygen losses (ROL) across the OPR suggest that, unlike water, rice roots efficiently retain oxygen within the aerenchyma. This ability strongly increases as roots/OPR develop. Key words: Aerenchyma, apoplastic transport, bulk flow, diffusional water permeabilty, exodermis, hydraulic conductivity, rice root, water channels.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Water transport within plants can be divided into several discrete steps, one of which is the radial water flow from the soil solution to root xylem vessels. It is known that radial water flow across roots is highly variable. This is due to a considerable variability in root hydraulic conductivity, which differs between plant species and in response to environmental conditions (Brouwer, 1954; Weatherley, 1982; Kramer and Boyer, 1995; Steudle, 2000a, b, 2001). Growth conditions strongly influence the architecture of roots including their anatomy and morphology (Steudle and Peterson, 1998; Steudle, 2000a). Physical and physiological processes can regulate water uptake by roots as well (Steudle, 2000b, 2001). In the longer term (days, weeks), the capacity to take up water can be related to root growth or structural, morphological, and anatomical changes of roots (i.e. the development of apoplastic barriers). In the short term (<1 d), root hydraulics are adjusted or are even regulated by physical properties, such as switching between cell-to-cell and apoplastic pathways. Cell-to-cell water flow is regulated by the gating of water channels (aquaporins) in the plasma membrane (Henzler and Steudle, 1995; Tyerman et al., 1999; Steudle, 2001; Javot and Maurel, 2002). The well-established composite transport model has been used to explain the variable permeability of roots to water (Steudle and Frensch, 1996; Steudle and Peterson, 1998; Steudle 2000a, b, 2001). The switching of water pathways may depend on both the driving forces and the water permeability of components of the pathway. This may allow some flexibility in the response of plants to water shortage according to the demand from the shoot.
Even for rice crops in paddy fields, water shortage may occur during the day with the appearance of wilting symptoms (Hirasawa et al., 1992, 1996). The hydraulic conductivity of rice roots is lower than other crop species because of a lack of flexibility in adjusting to demand from the shoot (Miyamoto et al., 2001; Ranathunge et al., 2003). The apoplastic barriers in rice roots, such as well developed exo- and endodermis with Casparian bands and lignified sclerenchyma cells, may restrict water movement through cell walls (Clark and Harris, 1981; Miyamoto et al., 2001; Ranathunge et al., 2003). The aerenchyma may represent an additional barrier. A comparison of detailed measurements of the overall hydraulic conductivity of rice roots (Lpr) with that of the outer part of roots (LpOPR) showed that the contribution of the endodermis/stele for the hydraulic resistance of rice roots was largest, followed by that of the aerenchyma (Ranathunge et al., 2003). Despite having the exodermis and sclerenchyma cells, the outer part of roots (OPR) had a hydraulic conductivity, which was larger by a factor of 30 than that of the whole roots or root systems (Ranathunge et al., 2003). Hence, water flow across the OPR must either have a strong apoplastic component or a prominent transmembrane passage of water flow or both in parallel. Low reflection coefficients of the OPR (sOPR) to non-permeating solutes such as mannitol pointed to a dominant apoplastic flow and a fairly porous apoplastic path rather than to a large transmembrane component of water flow across the OPR. These observations are in line with earlier findings of a substantial apoplastic transport of NaCl and the apoplastic dye trisodium, 3-hydroxy-5,8,10-pyrene trisulphonate (PTS) across the entire root cylinder of rice (Yeo et al., 1987; Yadav et al., 1996).
In this study, previous work on the hydraulic and osmotic properties of the OPR of rice roots has been extended in order to get an estimation of the contribution of pathways (apoplastic versus cell-to-cell components). Two different cultivars, a lowland (IR64) and an upland (Azucena) have been used again which may differ in their ability to take up water (root Lpr). Even though both pathways in roots contribute to the overall high water permeability across the OPR, it was found that the contribution of the apoplast was larger than that of the transmembrane passage of water flow favouring the presence of significant apoplastic bypasses. This was also suggested from experiments in which either the apoplastic passage was partially blocked using the suspension of China ink particles of an average diameter of 50 nm, or the membrane permeability was reduced using the water channel blocker HgCl2. Treatments created a significant reduction in LpOPR, but the reduction caused by blockage of the apoplast by ink particles was larger by a factor of 3 than that following HgCl2 treatment. Blockage caused a much bigger increase in the reflection coefficient of the OPR, when the apoplast was blocked. As expected in the presence of a fairly porous apoplastic bypass, there were huge differences between osmotic (bulk) and diffusional water permeabilities.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Plant material and growth conditions
Rice seedlings [Oryza sativa L. cv. Azucena (upland) and IR64 (lowland) from the International Rice Research Institute, Manila, Philippines] were grown from seeds in climatic chambers using aerated hydroponics as detailed previously (Ranathunge et al., 2003). Plants used in experiments were grown for 31–40 d including the time for germination.
Plant morphology and root anatomy
The height of young rice plants used in the experiments was 330–350 mm and 480–515 mm (eighth to eleventh leaf emergence) for IR64 and Azucena, respectively. Root lengths were typically 290–360 mm (IR64) and 500–550 mm (Azucena). Free-hand cross-sections were taken at distances of 20, 50, and 100 mm from the root apex and stained with Sudan Red 7B at room temperature for 1.5 h (Brundrett et al., 1991) to observe the development of aerenchyma along the roots. Sections were viewed using an optical microscope (DIALUX 22 EB, Leitz, Germany) and photographed using Kodak Elite 64 ASA film. To confirm the presence of suberin lamellae in the exodermis, free-hand cross-sections were stained for 1 h with Fluorol Yellow 088 (Brundrett et al., 1991) and viewed under an epifluorescence microscope using an ultraviolet filter set (excitation filter BP 365, dichroitic mirror FT 395, barrier filter LP 397; Zeiss, Oberkochen, Germany). To detect lignin in the cell walls of the OPR, free-hand cross-sections were stained for several minutes with phloroglucinol/hydrochloride at room temperature (Jensen, 1962). Lignin stained as a bright red layer within the cell walls.
Measurement of hydraulic conductivity of the outer part of roots (OPR) by pressure-perfusion
Root segments were excised 20–50 mm or 50–100 mm from the root apex. Aerenchyma was not fully developed at 20–50 mm from the apex, but it was at 50–100 mm. Both ends of each segment were fixed to glass capillaries with an inner diameter of 1.3 mm (Fig. 1A). One of the glass capillaries (inlet side) was connected to a syringe while the other (outlet side) was connected to a pressure probe by a narrow, rigid Teflon tube (inner diameter: 1.5 mm). The syringe was mounted on a 12-step Braun–Melsungen pump that produced defined pumping rates (water flows; QV) between 1.7x10–9 and 1.1x10–7 mm3 s–1. Perfusion from the inlet side of the root segments was commenced. Aerated nutrient solution was perfused through the aerenchyma displacing air. At a given pump rate, nutrient solution was pumped into the root segment and the pressure increased gradually until a stationary positive pressure was established where the volume flow produced by the pump equalled the radial volume flow across the OPR (Fig. 1B). The resulting steady-state pressures were measured using a pressure probe as a manometer. Stationary pressures were measured with respect to the flow rates. The hydraulic conductivity of the outer part of the root (LpOPR in m s–1 MPa–1) was calculated according to equation 1:
|
QV=LpOPRxPxAr(1)
where QV is the pump rate in m3 s–1, P is the steady-state pressure in MPa (reference: atmospheric pressure), and Ar is the surface area of the root segment (m2). To avoid anaerobic conditions, root segments were placed in a small chamber and an air-saturated nutrient solution was continuously circulated around the roots. A polyacrylamide glue (UHU, Bühl, Germany) was used to fix the root segments to the glass capillaries. This firmly attached even wet tissues to glass successfully. To make the seal mechanically rigid, the glue was superimposed with a molten mixture of beeswax/collophony (1:3 (w/w); Zimmermann and Steudle, 1975). The tightness of the seal was tested at the end of the experiments by perfusion with nutrient solution containing the apoplastic fluorescence dye 0.02% PTS (trisodium, 3-hydroxy-5,8,10-pyrene trisulphonate) as described previously (Ranathunge et al., 2003). In cases where the root segments were not sealed properly to the glass capillaries, the pressure-perfusion data were discarded. Given that the pressure-perfusion experiment lasted for 10–15 h, root segments were randomly selected for treatment with Evan’s Blue stain in order to test for the viability of the OPR cells (Fisher et al., 1985).
Osmotic experiments with perfused root segments
Reflection coefficients of the OPR (sOPR) were estimated by measuring changes in steady-state pressures caused by adding 14 mOsmol kg–1 mannitol (equivalent to 0.035 MPa of osmotic pressure) to the external medium, which did not change the hydraulic conductivity of the OPR (LpOPR). Hence, comparing the osmotic pressure applied with the change in hydraulic pressure gives the reflection coefficient (sOPR
–LpDOPR/LpOPR; LpDOPR=osmotic coefficient of the OPR). Mannitol was used as the test osmoticum since it has a of unity for plant cell membranes. The nutrient solution in the external medium was replaced with mannitol solution and the decline in steady-state pressures was observed. Maximum drops in pressures were used to calculate sOPR. Original steady-state pressures were obtained following the removal of mannitol from the external medium.
Blockage of cell-to-cell path of the OPR with water channel blocker HgCl2
50 µM HgCl2 was used as the blocking agent for cells of the OPR. After estimating the original steady-state pressures and reflection coefficients of the OPR (control), HgCl2 was added to the external medium of the root segments for 30 min, during which the direction of pumping of the perfusion pump was reversed (pump rate of 1.7x10–12 m3 s–1). This established a slightly negative pressure gradient across the OPR (–0.02 MPa relative to atmospheric pressure) drawing HgCl2 deeper into the tissue. Mercuric chloride could be also added to the perfusion medium. However, it was then difficult to get rid of it after the treatment. Excess HgCl2 was flushed from the system before new steady-state pressures and reflection coefficients were measured using the same flow rates as during the control. Following measurements, Hg2+ bound to the cell membranes of the OPR was scavenged by 4 mM 2-mercaptoethanol (Henzler and Steudle, 1995; Carvajal et al., 1996).
Blockage of porous apoplast by China ink particles
China ink particles (Rotring–Werke Riepe KG, Hamburg, Germany) were used to block apoplastic pores in the OPR in order to investigate the contribution of the apoplastic path to the overall water movement across the OPR. Prior to use, commercial China ink was diluted 1:1 with nutrient solution and cleared of small molecular weight compounds by dialysis against the nutrient solution. The osmotic concentration of the purified ink suspension was the same as that of the nutrient solution. The diameter of the ink particles was 51±22 nm, measured using a Particle Sizing System (PSS Nicomp, Santa Barbara, California, USA) by courtesy of Professor SD Tyerman, University of Adelaide, Australia. Root segments (20–50 mm or 50–100 mm from the root apex), previously used to measure the original steady-state pressures and reflection coefficients, were perfused with the diluted, purified China ink suspension for at least 1 h at relatively high flow rates directed from the inside to the outside, as described above, to block off pores in the apoplast (0.1 MPa pressure difference between aerenchyma and the medium). Following ink treatment, root segments were perfused with nutrient solution to sweep away excessive ink particles trapped inside the aerenchyma. As for the control, the resulting new steady-state pressures and reflection coefficients of the OPR were measured. The diluted, purified ink suspension was not toxic to cells of the OPR of rice roots.
Diffusional water permeability across the OPR with heavy water (HDO)
Steady-state perfusion using heavy water (HDO) was performed with root segments excised at either 20–50 or 50–100 mm from the root apex. A 3 M solution of HDO was perfused through the aerenchyma of root segments, displacing air with solution. The root segment was held vertically to allow perfusion of the solution by gravity (Fig. 2). The upper, open end of the segment (used as an inlet) was connected to a syringe by a Teflon tube filled with 3 M HDO. The other end of the root segment remained open as an outlet. As shown in Fig. 2, the syringe was placed 0.8 m above the root segment providing a gravitational force of 0.008 MPa, significantly less than in the previous pump-perfusion experiment (0.04–0.05 MPa; see above). Hence, water movement was near-isobaric (diffusive) to a good approximation and governed by the lateral diffusion of HDO across the OPR. Root segments were bathed in aerated nutrient solution of known volume (5 ml). At different time intervals, 50 µl of the outer medium was removed by a syringe and the HDO concentration of each sample was measured by a freezing point osmometer. The successive reduction of volume of the outer medium was accounted for. Since HDO and H2O form mixed crystals, the freezing point of samples containing HDO increased above that of distilled water in proportion to the concentration, as verified by a calibration curve (freezing points of distilled water: 0 °C; pure D2O: +4 °C). Measurements of diffusional water flow did not require root segments to be so tightly connected to the glass capillaries (inner diameter 1.3 mm) as during steady-state perfusion. Polyacrylamide glue was sufficient to connect wet tissue to glass. A small pump was employed to mix the external solution in order to equalize the distribution of HDO in the external medium and to minimize the thickness of the unstirred layers (Fig. 2).
|
In typical diffusional experiments, heavy water was perfused through the aerenchyma and diffused across the OPR to the external medium. The amount of the solute HDO that diffused to the outer medium was plotted against time. Solute flow across the OPR (JsOPR in mol s–1 m–2) was obtained directly from the slope of this curve divided by the surface area of the root segment. Since the external (diffused to outer medium) and internal (perfused through aerenchyma) HDO concentrations were known, the driving force or concentration difference between the inner and outer compartments (
Cs in mol s–1) could be evaluated. The diffusional water permeability of the OPR (PdOPR in m s–1) was obtained according to equation 2:
The external concentration was usually much smaller than the internal concentration, i.e. the back flow of HDO could be neglected. In order to compare the bulk/hydraulic water permeability (PfOPR) across the OPR with diffusional water permeability (PdOPR), the hydraulic conductivity of the OPR (LpOPR) was converted to PfOPR (House, 1974):
Here, 
w is the partial molar volume of liquid water. Since the unit of the PfOPR was m s–1, the PfOPR:PdOPR ratio was calculated directly. Following the diffusional permeability experiments, root segments were taken from the chamber and exposed to steam for 20–30 s to kill part of the living cells of the OPR. Then the diffusion experiment was repeated with steam-treated root segments.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Root anatomy
No visible anatomical or developmental differences were observed in the adventitious roots of the rice cultivars IR64 (lowland) or Azucena (upland). Therefore, Fig. 3 only refers to cross-sections of the lowland variety, IR64. At 20 mm from the root apex, cortical cells had partially collapsed to form large, gas-filled spaces, commonly called aerenchyma (Fig. 3A). At 100 mm from the apex, development of aerenchyma was complete (Fig. 3B). The initiation of cortical cell collapse was observed 5–6 cell layers from the innermost unmodified cortical layer. The OPR was separated from the stele by the large gas-filled spaces of aerenchyma. The OPR comprised the rhizodermis, exodermis, sclerenchyma (fibre cells), and one unmodified cortical cell layer. The dense packing of the sclerenchyma cells with thick cell walls indicated some physical strength of the OPR of rice roots. The cytoplasm of sclerenchyma cells was rarely observed closer to the root base, suggesting that they were definitely dead. Diamond-shaped air spaces were visible between the exodermis and the rhizodermis of the OPR where they loosely connected to each other. The OPR was connected to the stele by 40 to 50 spokes, i.e. monolayers of cells, which separated the different voids of the aerenchyma.
|
Weakly developed suberin lamellae were observed in the exodermis at a distance of 20 mm from the root apex (Fig. 3C), however, lamellae were fully mature at 50 mm from the root apex (Fig. 3D). No suberin was detected in sclerenchyma cell walls, even in mature zones close to the root base. The blue fluorescence of the sclerenchyma and cortex was autofluorescence. Sections taken at 50–100 mm from the apex showed weak deposits of lignin in the external tangential cell walls of the sclerenchyma tissue (Fig. 3E). At a given distance from the root apex, more lignin was observed in the exodermis than in the sclerenchyma.
Pressure-perfusion experiments with rice root segments
Two to three hours after fixing the root segments to the pressure-perfusion pump, stable pressures were established. These long time intervals were caused by the large volume (as compared to the conducting area of root segments) of the system, including the syringe and tubing. Increasing pump rates, linearly increased steady-state pressures. Stepwise increasing the pump rate (QV in m3 s–1) and then decreasing it again resulted in the same pressure/flow curves and LpOPR. There was no hysteresis in QV(P) curves. Neither IR64 (lowland) nor Azucena (upland) showed significant differences in LpOPR measured over the first 100 mm from the root apex (Table 1). OPR was quite permeable to bulk water. At 20–50 mm from the root apex, LpOPR values were (0.98±0.30)x10–6 and (0.99±0.20)x10–6 m s–1 MPa–1 for IR64 and Azucena, respectively (n=9 root segments each). Similar values were found for root segments at 50–100 mm from the apex (Table 1).
|
When all data were pooled and compared, the addition of 50 µM HgCl2 to the external medium did not significantly affect LpOPR (t-test; P=0.05) because of a large variation between root segments. However, when the results were presented as ratios of treatment/control LpOPR, the addition of HgCl2 reduced LpOPR significantly, by 10% in both cultivars and at both distances from the root apex (Table 1; t-test, P=0.05).
Root segments which were perfused with China ink for 1 h, to block the porous apoplast of the OPR (Fig. 4), had visibly darker root surfaces than controls (Fig. 5). Treatment with China ink decreased radial water flow across the OPR by 25% at 20–50 mm from the root apex, giving rates of (0.75±0.42)x10–6 and (0.73±0.28)x10–6 m s–1 MPa–1 for IR64 and Azucena, respectively (Table 1; n=9 root segments). The reduction was 30% at 50–100 mm from the apex, giving rates of (0.44±0.28)x10–6 and (0.59±0.19)x10–6 m s–1 MPa–1 (n=9 root segments). Means of the treatment/control ratios of LpOPR of individual roots were significantly smaller than unity (t-test; P = 0.05).
|
|
Overall, the treatment/control ratios of LpOPR obtained from the China ink treatment were significantly smaller (the effect of China ink treatment was bigger) than the ratios obtained from HgCl2 treatment. This was true for both rice cultivars at 20–50 mm and 50–100 mm from the root apex (t-tests; P=0.05; Fig. 6). It may suggest that more water passed through the apoplast rather than crossing the transmembrane passage.
|
Osmotic experiments with the outer part of rice roots
Efflux of water from root segments was induced by adding 14 mOsmol kg–1 mannitol. The resulting outward water flow caused a decline in stationary pressures of root segments at constant LpOPR (see Materials and methods; Fig. 7A, D as controls). The original steady-state pressures were restored on removal of mannitol from the external medium.
|
Before adding 50 µM HgCl2 to the external medium, steady-state pressures of root segments were dropped to sub-atmospheric (slightly negative
–0.02 MPa with reference to atmospheric pressure) as shown in Fig. 7B. This resulted in drawing HgCl2 deeper into the tissues. Higher steady-state pressures were obtained following the HgCl2 treatment than the control at the same pump rate (Fig. 7C). Root segments treated with HgCl2 showed larger drops in steady-state pressures upon the addition of mannitol than controls (Fig. 7A, C). In root segments 20–50 mm from the root apex, the HgCl2 treatment increased the sOPR from 0.12±0.04 to 0.23±0.05 in IR64 and 0.10±0.02 to 0.20±0.04 in Azucena (t-test; P=0.05; n=6 root segments). There was a similar increase in sOPR for root segments 50–100 mm behind the root apex (Table 2). On average for both root segments, the addition of HgCl2 increased the sOPR by a factor of about two.
|
Root segments perfused with diluted China ink for 1 h, developed higher stationary pressures than controls. The decline in stationary pressures upon the addition of mannitol was larger than that of controls (Fig. 7E). At a distance of 20–50 mm from the root apex,
sOPR increased from 0.13±0.04 to 0.40±0.09 in IR64 and 0.11±0.03 to 0.27±0.04 in Azucena (Table 2; n=6 root segments). Similar results were obtained for root segments taken 50–100 mm from the root apex for both cultivars. Overall, China ink perfusion of root segments increased the sOPR by a factor of about three. The increase in sOPR due to the China ink treatment was significantly greater than that due to HgCl2 treatment (t-test; P=0.05).
Diffusional water permeability of the OPR with heavy water
Vertical perfusion of aerenchyma by near-isobaric heavy water (HDO) was performed with excised rice root segments 20–50 mm or 50–100 mm from the root apex. The amount of HDO diffused into the external medium increased with time (Fig. 8). At any time, the external concentration of HDO was substantially smaller than that of HDO perfused through the aerenchyma which was constant. There was virtually no back flow of HDO (equation 2). Killing the roots by exposing to steam for 30 s, doubled the radial diffusion of HDO across the OPR into the external medium for both immature (20–50 mm) and mature (50–100 mm) root segments for both cultivars used (t-test; P=0.05). The diffusional permeability of the OPR (PdOPR) was obtained as defined by equation 2. Since the concentration of HDO in the aerenchyma was much larger than that of the external medium, this concentration was used as the driving force in equation 2. The diffusional water permeability of the OPR significantly decreased along the root axis from apex to base (t-test; P=0.05; n=6–7 roots). PdOPR was larger by a factor of two to three in immature (20–50 mm) compared with mature (50–100 mm) root segments. The PdOPR of root segments 20–50 from the root apex were 3.5±0.5 and 3.0±1.6x10–7 m s–1 in IR64 and Azucena, respectively. At a distance of 50–100 mm from the apex, values were 1.4±0.8 and 1.0±1.6x10–7 m s–1 in IR64 and Azucena, respectively. Steam treatment of root segments increased PdOPR by a factor of about two for both cultivars at both distances from the root apex. Comparison of bulk and diffusional permeabilities showed that the hydraulic/bulk water permeability of the OPR (PfOPR) was 600 times larger than the diffusional water permeability (PdOPR) at 20–50 mm from the apex and 1200–1400 larger at 50–100 mm from the apex (Table 3).
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
The present study provides further evidence of the passage of water through both the apoplast and the cell-to-cell path of the OPR of young rice roots. The passage across the two parallel pathways has been partially inhibited by either affecting water channel activity with the blocker HgCl2 or by closing pores in the apoplast with ink particles. Blocking off the apoplast is not easy. To the authors’ knowledge, the technique used here is unique. Results suggest that both pathways contribute to the overall water flow. The contribution of the apoplast appeared to be bigger, although roots developed apoplastic barriers as revealed by anatomical studies (e.g. suberin lamellae, Casparian bands). Because it comprises just four cell layers, the OPR of rice is a useful and well-defined structure for studying the tissue transport of water in the presence of apoplastic barriers. The anatomy of the OPR can be easily characterized by observing the development of an exodermis and sclerenchyma and the deposition of apoplastic barriers in parallel with changes in transport. Current models of tissue (root) water transport may be applied (Steudle and Frensch, 1996; Steudle and Peterson, 1998; Steudle 2000a, 2001). Ranathunge et al. (2003) have shown that the apoplast contributes to most of the radial permeability for water of the LpOPR. This is supported by the current results.
The absolute figure of LpOPR was larger by a factor of 30 than the overall Lpr value of rice roots. It seems that at least these two rice cultivars differ from other crop plants where the contribution of the hydraulic resistance of the exodermis is much bigger. Maturation of the exodermis substantially reduced the radial water flow in onion (Melchior and Steudle, 1993). In young corn plants, the exodermis caused a 4-fold reduction of root Lpr (Zimmermann and Steudle, 1998).
Even for the same cultivar grown under similar conditions, detailed anatomical studies of the OPR of rice roots confirmed that exodermis maturation was highly variable, as previously shown by Perumalla and Peterson (1986) for onion and corn roots. This specialized type of hypodermis (exodermis) allowed a rather free flow of bulk water across the periphery of rice roots. Even though fully matured exodermis with Casparian bands was found beyond 60–80 mm from the root apex, this did not significantly reduce radial water flow across the OPR. This may be due to local disruption of the exodermis when developing lateral roots from the pericycle allowing high apoplastic bypasses through these cracks (Peterson et al., 1981). In roots lacking an exodermis, water and ions can potentially move apoplastically through the cell walls of the epidermis and cortex as far as the endodermis (Peterson, 1988). Despite having the exodermis, the OPR was reasonably permeable for water. Preliminary results showed that charged ions could also pass through premature Casparian bands in the rice exodermis, at least to some extent (data not shown). This is an agreement with earlier findings of Flowers and co-workers of an apoplastic bypass flow of sodium and the apoplastic tracer PTS (Yeo et al., 1987; Yadav et al., 1996).
It may be argued that the exodermis is permeable because of the patchiness of Casparian bands or because of the presence of passage cells lacking suberin lamellae. It is known that rice roots contain passage cells in the exodermis (Clark and Harris, 1981). However, in these histochemical studies it was not possible to confirm a large number of passage cells in the exodermis for the rice cultivars used here. There was no evidence for patchiness of the exodermis. On the contrary, maturation of Casparian bands and suberization of the hypodermis was fairly uniform along the roots for both cultivars. It was completed at a distance of 50–60 mm from the apex. It was known that the presence of suberin lamellae in hypodermal cell walls of corn sleeves did not necessarily indicate a low permeability to water or solutes (Clarkson et al., 1987). This is an agreement with the present data.
The permeability properties of the apoplastic barriers for water or solutes are related to the amount and chemical composition of aliphatic and aromatic suberin and lignin (Schreiber et al., 1999; Zimmermann et al., 2000; Hose et al., 2001). Neither uniseriate sclerenchymatous layers formed of short fibres (Clark and Harris, 1981) nor suberized and lignified mature exodermis greatly impeded the radial bulk flow of water across the OPR much. Histochemical studies with rice roots showed that lignification of the exodermis started as close as 50 mm from the root apex. No lignin was detected in sclerenchyma fibre cells at the same distance. According to Clark and Harris (1981), lignified sclerenchyma fibre cells without cytoplasm were observed as far as 150 mm from the root apex. Unusually low amounts of lignification of sclerenchyma fibre cells at distances of up to 100 mm may indicate a rather high permeability of this structure, which is largely composed of cellulose.
Basically, radial water flow across the roots could use either the apoplastic or cell-to-cell pathways or both. There is strong evidence that water channels (aquaporins) play a central role in plant water relations (Chrispeels and Maurel, 1994; Henzler and Steudle, 1995; Maurel, 1997; Steudle and Peterson, 1998; Steudle, 2001; Tyerman et al., 1999, 2002; Javot and Maurel, 2002). Water channel activity can be affected by different parameters such as high salinity, nutrient deprivation, drought, diurnal rhythms, and heavy metals (Azaizeh and Steudle, 1991; Carvajal et al., 1996, 1999, 2000; North and Nobel, 2000; Henzler et al., 1999; Henzler and Steudle, 1995). The exact mechanisms of the gating of channels are poorly understood (Tyerman et al., 1999, 2002; Steudle, 2000a, b, 2001; Ye et al., 2004; Wan et al., 2004). A tentative indicator of water channel involvement is the observation that heavy metals like Hg2+ can reversibly reduce the hydraulic conductivity of roots by binding to -SH groups of water channels (Henzler and Steudle, 1995; Maggio and Joly, 1995; Carvajal et al., 1996; Wan and Zwiazek, 1999; Barrowclough et al., 2000; North and Nobel, 2000). According to the authors’ best knowledge, this is the first study in which water channels were blocked off with HgCl2 for only a part of roots (just its outer part or periphery).
For Agave deserti, the reduction of radial hydraulic conductivity was 60% in the presence of 50 µM HgCl2 under wetted conditions (Martre et al., 2001). It was 4-fold in the basal root zones of onion (Barrowclough et al., 2000). These figures differ from the OPR of rice roots which showed only a 10% reduction. The main reason for that might be that apoplastic barriers like Casparian bands, suberin lamellae or lignin restricted the penetration of even non-dissociated HgCl2 into OPR and to plasma membranes. Perhaps, the reduction of the radial water flow across the OPR only resulted in a closure of water channels in the rhizodermis and external membranes of the exodermis. Alternatively, the water channels in the OPR may not contain many -SH groups or these groups are difficult to access with HgCl2 because of the suberization. Hence, roots were not particularly sensitive to HgCl2. Since the major barrier to radial water flow in rice roots was the endodermis (Miyamoto et al., 2001; Ranathunge et al., 2003), most of the water channels might be located around the endodermis as found for other species (Schäffner, 1998).
Blockage of the apoplastic pathway by ink treatment reduced LpOPR by 30% which was significantly larger than the 10% inhibition caused by HgCl2. Perhaps, the ink suspension used could not effectively close all pores in the cell walls because of the relatively large particle sizes. The mean was 51±22 nm which is bigger than the diameter of interfibrillar pores (interstices) which are about 5–30 nm (Nobel, 1999). Partial blockage of the apoplast in the OPR reduced radial water flow by more than the water channel blocker HgCl2, suggesting relatively large apoplastic bypasses. There is a need for better apoplastic blockers. Tests are underway with suspensions of particles of smaller mean diameter, which should result in a larger reduction of LpOPR of, say, by a factor of 5–10. Completeness of blockage may be tested by measuring the reflection coefficient, which should increase when blockage of the apoplast is complete or nearly so.
The picture of a dominating apoplastic rather than cell-to-cell path for radial water flow within the OPR is in line with the low overall reflection coefficients (sOPR 0.1; Table 2; Ranathunge et al., 2003) as well as the doubling of sOPR upon partial pore closure by ink particles. The solute mannitol used to measure effects on sOPR does not permeate plant cell membranes and should have a scc 1 along the cell-to-cell path (nearly semipermeable membranes). Unlike membranes, however, the apoplast should have scw 0 with virtually no selectivity expected (Steudle, 2000a, b). Roots, having complex structures with the two pathways arranged in parallel as well as in series, then overall s usually locates between 0 and 1, which can be calculated using a relation derived from basic irreversible thermodynamics (Kedem and Katchalsky, 1963a, b). Partial closure of apoplastic pores with ink resulted in a decrease of the movement of solutes (mannitol) through the apoplast leading to a higher sOPR than the treatment with the water channel blocker HgCl2.
According to the composite transport model, which may be applied for complex structures such as roots, the overall reflection coefficient (s) for a parallel arrangement of membranes should decrease after closing water channels with HgCl2 (assuming that scc 1 and scw 0). Data presented in this paper showed the opposite trend: s increased after HgCl2 treatment. The OPR of rice roots contain four cell layers in series, indicating that both parallel and serial membrane models contribute to the overall reflection coefficient. It is not clear why this deviation occurred but it may arise because the OPR comprises both parallel (apoplast versus cell-to-cell) and four different series layers of cells, which may differ in their transport properties. For example, the Kedem and Katchalsky (1963b) treatment of patchy membrane systems predicts that, in a series array, the overall reflection coefficient would be the weighed sum of individual arrays, whereby the series elements contribute according to their solute (mannitol) permeability. It can not be excluded that the permeability of mannitol of different layers is affected by HgCl2 treatment, although direct evidence is missing.
In order to compare the LpOPR (units: m s–1 MPa–1) with the osmotic water permeability, Pf in units of m s–1, equation (3) was used. Pf rather than Lp is usually given in animal physiology (e.g. Table 5.6 in House, 1974). The diffusional permeability of the OPR of rice roots (PdOPR) was much lower than the osmotic water permeability (PfOPR). Absolute values of PdOPR were bigger than those of sleeves of the aerenchymatous species Carex arenaria (10–8 m s–1; Robards et al., 1979), but smaller than the PdOPR of sleeves obtained from aerenchymatous corn roots (10–6–10–7 m s–1, depending on the position from the root apex; Clarkson et al., 1987). By a factor of as large as 600–1400, the osmotic water permeability (PfOPR) was greater than that of diffusional water permeability (PdOPR). These values were larger than the Pf/Pd ratio of artificial membranes (Pf/Pd=1–730; Table 4.4 in House, 1974) and various animal tissue (Pf/Pd = 1 to 300; Table 9.5 in House, 1974). It should be noted that, in the present experiments, diffusional water flows were measured under near-isobaric but not completely isobaric conditions. Hence, Pd may be overestimated, and the large ratios represent a lower limit. Such large Pf/Pd ratios are expected if the pathway involved a rather long porous path; this would offer a high diffusional resistance for HDO, but should be highly permeable in case of a bulk (hydraulic) water flow. In single-file pores such as water channels, the ratios of Pf/Pd >1 are a measure of the number of water molecules aligned within the pore (Levitt, 1974). It is well documented that Pf/Pd ratios may be overestimated in the presence of unstirred layers, which affect Pd rather than Pf. However, during the measurements in this study, the solutions in both compartments (aerenchyma and external solution) were well-stirred, tending to reduce the effect. Hence, large ratios were not due to the effects of unstirred layers. In the context of the other findings of (i) effects of blocking experiments and (ii) low reflection coefficients, the huge Pf/Pd ratios provide the strongest evidence for a major passage of water along the apoplast, even in the presence of apoplastic barriers.
Rice roots often grow under water-logged conditions in hypoxic soil environments. At first sight, a low diffusional permeability (as found for water) may also refer to oxygen, which diffuses from the shoot to the root tips through the aerenchyma under hypoxic conditions (Armstrong, 1979; Colmer et al., 1998). To reach the root tips, it is required that there are no excessive losses to the soil, i.e. PO2 should be low. The differences between diffusional (HDO) and bulk water (Lp) permeabilities indicate that this could be achieved by differences in the transport mechanism (diffusional versus bulk flow; Ranathunge et al., 2003). Hence, rice roots could have a rather high bulk water permeability in the presence of a low permeability to oxygen, which reduces radial oxygen losses (ROL). This would be favourable to the plant. The present data show that the diffusional water permeability was only reduced by a factor of 3 as roots developed. At the same time, radial oxygen loss (ROL) drastically decreased along the roots and ended up with rates of close to zero at a distance of 50 mm from the root apex for the cultivars used (L Kotula, K Ranathunge, E Steudle, R Lafitte, unpublished data). Apparently, the diffusion of oxygen from the aerenchyma to the outer medium is strongly restricted by the existence of apoplastic barriers, which retain oxygen more effectively than water. This may point to differences in the transport path for the two compounds. However, there are, to date, no data of the permeability coefficients of oxygen across the OPR to compare with the permeabilities of water and how this would change during root development. These values are badly needed.
In conclusion, the data show that apoplastic water flow contributes much more to the overall water flow across the OPR of rice roots than the transmembrane component. The findings suggest that exodermal apoplastic barriers such as Casparian bands and suberin lamellae are fairly permeable to water. Partial blockage of the porous apoplast with ink particles proportionately reduced the radial water flow across the OPR more than HgCl2 did along the cell-to-cell path, suggesting that there were prominent apoplastic bypasses. This was in line with substantial relative increases of sOPR in response to the blockage of the apoplast with ink rather than the cell-to-cell path. However, absolute values of reflection coefficients remained rather low. The diffusional water permeability (PdOPR) was smaller by two (for immature root segments) or three (for mature root segments) orders of magnitude than the osmotic (PfOPR). This strongly supported the view that there was substantial apoplastic transport of water across the OPR of rice, even in the presence of Casparian bands and suberin lamellae. Diffusional and bulk water permeabilities did not decrease much during root development. Hence, the small effect of root development on the diffusional permeability of water differed from that found for oxygen (Colmer et al., 1998). This suggested that the two diffusants use different pathways within the OPR.
|
Acknowledgements |
|---|
We thank Ms Alaina Garthwaite, University of Western Australia, Perth for carefully reading and correcting the manuscript. We wish to thank two anonymous refrees for helpful comments and suggestions. The excellent technical assistance of Burkhard Stumpf (Lehrstuhl Pflanzenökologie, University of Bayreuth) is gratefully acknowledged. We are also grateful to the Deutsche Forschungsgemeinschaft: Ste319/3-3 (ES), and BMZ (project No. 2000.7860.0-001.00) ‘Trait and Gene Discovery to Stabilize Rice Yields in Drought Prone Environments’ (RL) for financial support. LK thanks for a grant within the ERASMUS/SOCRATES programme of the EU.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Armstrong W. 1979. Aeration in higher plants. Advances in Botanical Research 7, 225–332.
Azaizeh H, Steudle E. 1991. Effects of salinity on water transport of excised maize (Zea mays L.) roots. Plant Physiology 97, 1136–1145.
Barrowclough DE, Peterson CA, Steudle E. 2000. Radial hydraulic conductivity along developing onion roots. Journal of Experimental Botany 51, 547–557.
Brouwer R. 1954. The regulating influence of transpiration and suction tension on the water and salt uptake by roots of intact Vicia faba plants. Acta Botanica Neerlandia 3, 264–312.
Brundrett MC, Kendrick B, Peterson CA. 1991. Efficient lipid staining in plant material with Sudan red 7B or Fluorol yellow 088 in polyethylene glycol-glycerol. Biotechnic and Histochemistry 66, 111–116.
Carvajal M, Cerda A, Martinez V. 2000. Does calcium ameliorate the negative effect of NaCl on melon root water transport by regulating aquaporin activity? New Phytologist 145, 439–447.[CrossRef]
Carvajal M, Cooke DT, Clarkson DT. 1996. Response of wheat plants to nutrient deprivation may involve the regulation of water-channel function. Planta 199, 372–381.
Carvajal M, Martinez V, Alcaraz CF. 1999. Physiological function of water channels as affected by salinity in roots of paprika pepper. Physiologia Plantarum 105, 95–101.[CrossRef]
Chrispeels MJ, Maurel C. 1994. Aquaporins: the molecular basis of facilitated water movement through living plant cells. Plant Physiology 105, 9–15.[CrossRef][ISI][Medline]
Clark LH, Harris WH. 1981. Observation on the root anatomy of rice (Oryza sativa L.). American Journal of Botany 68, 154–161.[CrossRef]
Clarkson DT, Robards AW, Stephens JE, Stark M. 1987. Suberin lamellae in the hypodermis of maize (Zea mays) roots; development and factors affecting the permeability of hypodermal layers. Plant, Cell and Environment 10, 83–93.
Colmer TD, Gibbered MR, Wiengweera A, Thin TK. 1998. The barrier to radial oxygen loss from roots of rice (Oryza sativa L.) is induced by growth in stagnant solution. Journal of Experimental Botany 49, 1431–1436.
Fisher JMC, Peterson CA, Bols NC. 1985. A new fluorescent test for cell vitality using Calcofluor white M2R. Stain Technology 60, 69–79.[ISI][Medline]
Henzler T, Steudle E. 1995. Reversible closing of water channels in Chara internodes provides evidence for a composite transport model of the plasma membrane. Journal of Experimental Botany 46, 199–209.
Henzler T, Waterhouse RN, Smyth AJ, Carvajal M, Cooke DT, Schaffner AR, Steudle E, Clarkson DT. 1999. Diurnal variations in hydraulic conductivity and root pressure can be correlated with the expression of putative auaporins in the roots of Lotus japonicus. Planta 210, 50–60.[CrossRef][ISI][Medline]
Hirasawa T, Tenmyo N, Suyuki M, Ishihara K. 1996. Mechanisms of occurrence of white head of rice plants at heading stage under high temperature, low humidity and high wind velocity conditions: factors increasing stem resistance to water flow at the panicle base. Japanese Journal of Crop Science 65, (Extra issue 2) 129–130.
Hirasawa T, Tsuchida M, Ishihara K. 1992. Relationship between resistance to water transport and exudation rate and the efect of the resistance on the midday depression of stomatal aperture in rice plants. Japanese Journal of Crop Science 61, 145–152.
Hose E, Clarkson DT, Steudle E, Schreiber L, Hartung W. 2001. The exodermis: a variable apoplastic barrier. Journal of Experimental Botany 52, 2245–2264.
House CR. 1974. Water transport in cells and tissues. London: Edward Arnold.
Javot H, Maurel C. 2002. The role of auaporins in root water uptake. Annals of Botany 90, 301–313.
Jensen WA. 1962. Botanical histochemistry. San Francisco: WH Freeman and Company.
Kedem O, Katchalsky A. 1963a. Permeability of composite membranes. Part 2. Parallel arrays of elements. Transactions of the Faraday Society 59, 1931–1940.
Kedem O, Katchalsky A. 1963b. Permeability of composite membranes. Part 3. Series array of elements. Transactions of the Faraday Society 59, 1941–1953.
Kramer PJ, Boyer JS. 1995. Water relations of plants and soil. Orlando: Academic Press.
Levitt DG. 1974. A new theory of transport for cell membrane pores. I. General theory and application to red cell. Biochimica et Biophysica Acta 373, 115–131.[Medline]
Maggio A, Joly RJ. 1995. Effects of mercuric chloride on the hydraulic conductivity of tomolo root systems: evidence for a channel-mediated water pathway. Plant Physiology 109, 331–335.[Abstract]
Martre P, North GB, Nobel PS. 2001. Hydraulic conductance and mercury-sensitive water transport for roots of Opuntia acanthocarpa in relation to soil drying and rewetting. Plant Physiology 126, 352–362.
Maurel C. 1997. Aquaporins and water permeability of plant membranes. Annual Review of Plant Physiology and Plant Molecular Biology 48, 399–429.[CrossRef][ISI][Medline]
Melchior W, Steudle E. 1993. Water transport in onion (Allium cepa L.) roots. Plant Physiology 101, 1305–1315.[Abstract]
Miyamoto N, Steudle E, Hirasawa T, Lafitte R. 2001. Hydraulic conductivity of rice roots. Journal of Experimental Botany 52, 1–12.
Nobel PS. 1999. Physicochemical and environmental plant physiology. San Diego: Academic Press Inc.
North GB, Nobel PS. 2000. Heterogeneity in water availability alters cellular development and hydraulic conductivity along roots of a desert succulent. Annals of Botany 85, 247–255.
Perumalla CJ, Peterson CA. 1986. Deposition of Casparian bands and suberin lamellae in the exodermis and endodermis of young corn and onion roots. Canadian Journal of Botany 64, 1873–1878.
Peterson CA. 1988. Exodermal Casparian bands, their significance for ion uptake by roots. Physiologia Plantarum 72, 204–208.
Peterson CA, Emanuel ME, Humphreys GB. 1981. Pathway of movement of apoplastic dye tracers through the endodermis at the site of secondary root formation in corn (Zea mays) and broad bean (Vicia faba). Canadian Journal of Botany 59, 618–625.
Ranathunge K, Steudle E, Lafitte R. 2003. Control of water uptake by rice (Oryza sativa L.): role of the outer part of the root. Planta 217, 193–205.[ISI][Medline]
Robards AW, Clarkson DT, Sanderson J. 1979. Structure and permeability of the epidermal/hypodermal layers of the sand sedge (Carex arenaria L.). Protoplasma 77, 291–312.[CrossRef]