Journal of Experimental Botany, Vol. 48, No. 310, pp. 985-999, May 1997
Journal of
Experimental
Botany
OPINION PAPER
Primary responses of root and leaf elongation to water
deficits in the atmosphere and soil solution
Jiirgen Frensch1
Lehrstuhl fOr Pflanzendkologie, Universita't Bayreuth, D-95440 Bayreuth, Germany
Received 5 August 1996; Accepted 16 January 1997
Abstract
Plants experience drought by a limitation of water
supply and by enhanced transpiration. Both processes
tend to decrease the plant's water potential, but affect
growth responses in the root and leaf differently. The
evaluation of the underlying mechanisms leads to a
discussion of recent studies on biophysical aspects of
cell expansion at a cellular, tissue and organ level.
Two processes enable roots to compensate rapidly
effects of water deficits originating in the medium: (i)
adjustment of the minimum pressure in cells required
for expansion (yield threshold), and (ii) solute transport
within the elongation zone. Limitations of root growth
are discussed with respect to hydraulic, mechanical,
and solute relations in the root elongation zone. It is
argued that the variable nature of both the yield
threshold and solute transport challenges the applicability of the Lockhart concept to determine growthrelated parameters from steady conditions of turgor
and growth. On a whole organ level, the attenuation
of xylem pressure along the root is important for
the differential response of root and leaf growth.
Experimental evidence is presented for the hydraulic
separation of the elongation zones, which is closely
related to root development and functioning. The data
obtained over the past few years have been used to
extend mathematical models of growth and water
transport in roots.
Key words: Extension growth, hydraulic conductivity, root
development (xylem, endodermis), transport (water and
solute), turgor pressure, water stress, xylem pressure,
Zea mays.
Introduction
At least during short periods of time, terrestrial plants
are exposed to water limiting conditions in almost every
1
Fax: +49 921 552564. E-mail: jurgen.frenschQunl-bayreuth.de
C Oxford University Press 1997
natural environment. Cell expansion is among the most
sensitive processes in plants affected by drought which is
the result of an imbalance in the plant between water
uptake and water loss (Hsiao, 1973; Boyer, 1985). Water
loss to the atmosphere is largely controlled by stomatal
movement. Stomata respond to drought in the atmosphere and in the soil, and much information on the
underlying mechanisms has been accumulated (Schulze,
1986; Tardieu and Davies, 1993; Slovik et al., 1995). By
contrast, the complexity between root performance and
low plant water potentials is much less understood. To
maintain uptake of water and nutrients under water
limiting conditions, continuous proliferation of roots into
new soil layers is important and requires adaptive mechanisms on the cell and tissue level (Tomos et al., 1989;
Carpita and Gibeaut, 1993; Cosgrove, 1993).
This review aims to evaluate limitations of root and
leaf growth effected by drought in the atmosphere and in
the soil. In particular, short-term responses of hydraulic
conductivity, cell wall yielding, and solute transport in
the root elongation zone are analysed. Furthermore, the
significance of root development for the differential
response of roots and shoots to drought stress is emphasized and demonstrated. Mechanisms of long-distance
volume flow are discussed with respect to new experimental results on pressure propagation in the xylem.
Current mathematical models of cell expansion and water
uptake are extended on the basis of experimental data.
Differential response of the root and the leaf to
water stress: Lockhart's model
For many plants, root growth is more resistant to water
stress than shoot growth (Hsiao and Jing, 1987; Kramer
and Boyer, 1995). The different responses may be exemplified by a simple experiment. Root and leaf elongation
were measured continuously on the same intact plant
986 Frensch
which was exposed to various treatments affecting its
water status. At steady rates of root and leaf elongation,
the seedlings were subjected to an osmotic step change in
the medium (Fig. 1A). The osmoticum lowered the water
potential of the medium and reduced the water availability
instantaneously. Effects of the osmoticum (mannitol) on
growth other than a change in water potential were
negligible during the short-term period of exposure of
1-2 h (Frensch and Hsiao, 1994). It can be seen in Fig. 1
that root elongation stopped almost immediately upon
addition of the osmoticum, but resumed after a few min
at a new rate smaller than prior to the treatment. The
response of leaf elongation was different in that growth
remained stronger inhibited than in the root, at least
within the first h of exposure to water stress.
When transpiration was increased by a step change of
the humidity inside the gas exchange cuvette a different
type of growth response was observed (Fig. IB). Leaf
elongation declined significantly, but root elongation was
essentially unaffected. The contrasting behaviour of the
two organs indicated that transpiration competed for
water with leaf growth rather than with root growth. The
fact that growth was less affected in the root than in the
leaf may be related to different mechanical and hydraulic
properties in these organs (Hsiao and Jing, 1987). An
alternative explanation of the differential growth involves
the hydraulics of the plant, particularly that of the root.
Both issues will be addressed below in detail.
Over the last three decades, the mechanical and
hydraulic processes of cell expansion have been analysed using Lockhart's framework (Lockhart, 1965).
Accordingly, the irreversible increase of cell size (dVjdt)
is associated with a relaxation of the cell wall network,
which is driven by a turgor pressure (Pc) in excess of a
critical threshold (Y). A steady cell enlargement is
achieved when water uptake just compensates turgor
reduction caused by the volume increase of the protoplast.
Under these conditions, the governing equation for steady
growth of a single cell is:
(
dt \mV
+
(i)
LpA
where m is the volumetric extensibility coefficient of the
3 leaf LVDT
gas exchange
cuvette
leaf elongation
zone
nutrient
solution or
osmoticum
root LVDT
root elongation
zone
root support
11.0
to pump
12.0
11.5
12.5
Time (h)
Fig. 1. Simultaneous measurement of root and leaf elongation of a young maize plant. Trie environment of the shoot was controlled by a gas
exchange cuvette; a flowing solution culture covered the primary root clamped to a temperature-controlled root support. Two position transducers
(LVDTs) attached to the first leaf and to the root monitored the plant's response to changes in water potential effected by (A) osmotic step
changes in the medium and (B) increased VPD (decreased dew point temperature, 7"dp). Depending on the nature of the treatment the experiments
yielded different types of growth responses in roots and leaves. While a restriction of the water supply affected both organs, although differently
during the transition, root growth was largely unaffected upon increasing transpirational demand. Total root length: 150 mm; plant age: 4 d after
germination and transfer to growth containers; photosynthetic active radiation at the top of the chamber: approximately 1000 ^unol m~2 s"1; air
temperature: 24 26 C; nutrient solution similar to a quarter-strength Johnson solution.
Roof elongation and drought
wall. Lp is the hydraulic conductivity of the cell, A and
Fare the surface area and volume of the cell, respectively,
nc is the osmotic pressure of the cell sap, and y0 is the
water potential of the source. The two terms in brackets
represent mechanical and hydraulic 'growth resistances'
which limit cell expansion according to their contribution
to the overall resistance.
Provided that the mechanical resistance is much larger
than the hydraulic, Equation (1) reduces to
]
dt \mV
P
'
Y
(2)
where (Pc-Y) denotes the 'growth-effective turgor'.
Equation (2) rather than equation (1) is usually used to
quantify cell expansion. When solute transport into the
cell is not unlimited, however, the simplified equation can
be misleading. Because TTC is proportional to the ratio of
volume and solute flows (Jy/J,), the criteria for the
simplification are not justified solely by showing that
1/fmx V)»\/(LpxA) and, thus, Wos Wm (fm: medium
water potential). Rather the relationship between -nc and
J, needs further consideration, which is largely ignored
in studies relating to cell and tissue growth. It is expected
that a separation of solute and water flows would help
to elucidate the controversial debate on the significance
of hydraulic conductivities for growth.
Conceptually, the differential response of growth in
roots and shoots to drought may be summarized in terms
of different control loops, linking cell expansion to biophysical and biochemical processes in growing tissues
(Bradford and Hsiao, 1982). Turgor pressure (Pc) has
long been recognized a driving force for cell enlargement,
although its meaning for growth control and adjustment
to water stress is debated. Doubts about Pc as a limiting
factor are based on results from experiments in the field
and laboratory (Matsuda and Riazi, 1981; Michelena and
Boyer, 1982; Shackel et al, 1987; Hsiao and Jing, 1987;
Munns, 1988). However, growth may be sensitive to
changes in Pc only where expansion is limited or co-limited
by low turgor. This hypothesis was recently investigated
by exposing maize plants to changes in water potentials
effected in the root medium (osmotic and hydrostatic
pressure steps) and in the atmosphere (humidity steps)
(Frensch, Hsiao, Rochas-Lara, unpublished results).
The response of the first maize leaf to water deficit was
complex and depended on environmental and developmental conditions of the plant (Fig. 2). Highest rates of
leaf elongation were obtained when ambient air humidity
was high and Wm was close to zero. Any further increase
of xylem water potential mediated by small hydrostatic
pressure steps in the medium (0.025 MPa) failed to
increase the leaf elongation rate (Fig. 2, open triangles).
Apparently, water was not limiting extension, and sensitivity of growth to changes in Pc was low. With successively
1
1
1
1
987
'
0.50 -
>
*
CD
2
0 25 g
T
/
~o
CD
O)
CO
0.001
6
0
1
•
1
20
40
Length of first leaf (mm)
•
60
Fig. 2. Sensitivity of leaf growth (first maize leaf) to small hydrostatic
pressure steps of 0.025 MPa (25 kPa) in the xylem at different stages of
development. Plants were subjected to drought stress treatments in the
medium and atmosphere At their maximum water status (open
triangles; low transpiration, unrestricted water supply) plants did not
respond to a xylem pressure step which was induced by pressurizing
the root medium. When the water potential of the medium was lowered
to - 0 . 3 MPa (mannitol) and transpiration was kept low (open circles),
pressure steps altered the elongation rate significantly in leaves larger
than 30 mm. A further increase in sensitivity of leaf growth to
hydrostatic step changes was observed when transpiration was increased
in the presence of a 0.3 MPa mannitol solution (closed circles). The
results indicate that the sensitivity of leaf growth to changes in xylem
pressure depends on environmental and developmental conditions. Data
are means ±SE. Data points indicated by asterisks indicate a significant
difference between the treatments presented by open and closed symbols
(f-test, /> = 0 05) (Unpublished data, Frensch, Hsiao, Rochas-Lara).
less favorable growth conditions (increase of transpiration, *F m =-0.3MPa), leaf elongation benefited from
the raise of xylem water potential and additional supply
of water. Presumably, the increase of xylem pressure
reduced the water potential gradient between the xylem
and the surrounding elongating tissue (Nonami and
Boyer, 1993). The rather small pressure changes involved
illustrate the high sensitivity of leaf elongation to the
plant water status when growth was competing with
transpiration for water.
A high sensitivity of growth to similar small pressure
steps has been previously reported for hypocotyl segments
of Vigna (Okamoto et al, 1989). Different from maize,
hypocotyls responded transiently to pressure changes
which were applied to the solution perfusing the hollow
segments. The transient behaviour of hypocotyl growth
correlated well with the adjustment of wall rheology.
Additional experiments yielded a close correlation
between growth adjustment and the acidification of the
cell wall solution (Mizuno et al, 1993; Okamoto and
Okamoto, 1994). These studies provide evidence for the
close coupling of metabolic events and biophysical
changes of growth parameters.
Limitations of Lockhart's model
Lockhart's model has been proven to be very useful to
quantify the growth process of a single cell. None the
988 Frensch
less, there are obvious limitations of the model when
applied to a growing tissue, which have not yet been
incorporated in the theoretical framework. Tissue growth
differs from cell growth in that:
(a) growth rates vary along the elongation zones in the
plant, corresponding to (i) a spatial heterogeneity in the
demand for water and solutes, (ii) variable growth-related
parameters with respect to position and time, and
(b) water and solutes are supplied by different compartments (e.g. xylem, phloem, soil solution) and transport
involves parallel pathways (apoplast and symplast), each
with distinct physical and chemical properties.
Measurements at a high spatial and temporal resolution
are required to unravel causes and effects within the
complex process of tissue growth, and to evaluate tissue
response to environmental factors properly.
It is important to recall that equation (2) would predict
a linear relationship between Pc and growth rate. Changes
of Pc are commonly affected by altering the rate of water
uptake by means of changes in Wo, nc, and LpxA.
Following perturbations of Pc, linearity would justify that
Y and m can be determined from steady growth rates
and steady turgor pressures. Despite early concerns
regarding the linearity (Acevedo et al., 1971; Green et al.,
1971) this approach has been extensively used to determine Y and m. Alternatively, Y has been determined from
turgor relaxation after water uptake into the cell was
inhibited (Ortega, 1985; Cosgrove, 1987). However, continuous measurements of Pc during transient responses of
growth rates indicate that cell wall parameters are variable
and adjust in a matter of min to new environmental
conditions (Shackel et al., 1987; Serpe and Matthews,
1992; Frensch and Hsiao, 1994). Without adequate considerations of the variability, previous methods may lead
to erroneous values of Y and m. Until now, Lockhart's
model does not account for the variable nature of any
parameter in equation (1). A significant step toward this
direction is the modelling of responses of microfibrils in
the cell wall to changes in turgor (Passioura and Fry,
1992). Further measurements are required to relate metabolic events with physical restrictions or enhancements of
cell expansion.
The role of solute accumulation for growth
Water uptake and volume increase of the growing cell
constantly dilute the cell sap which would lead to turgor
relaxation until growth ceases at Pc= Y. Thus, cell expansion is intimately related to solute uptake and/or synthesis.
To some extent, equation (1) accounts for this relationship
by the parameter nc. Introduced as part of the driving
force for water flow, nc is also a measure of the solute
concentration in the cell. It is evident, however, that
equation (1) does not govern the coupling of water and
solute flows during cell enlargement. This may be possible
by the introduction of a third 'growth resistance' that
would couple solute flow to an appropriate force. For
example, active solute uptake may be related to the
metabolic energy of the cell, whereas passive solute
movement into the cell may be driven by a concentration gradient between adjacent cells connected by
plasmodesmata.
Depending on the location, the various elongation
zones of a plant gain solutes from different compartments.
For the root elongation zone, solute accumulation is
probably sustained by two compartments, i.e. the soil
solution (nutrient salts) and the phloem (carbohydrates,
nutrient salts). For above-ground organs, phloem and
xylem supply growing cells. To measure rates of solute
accumulation in a single cell, consider the following
experiment. Turgor is lowered to a level well below Y,
and cell expansion stops. As long as Pc < Y, the irreversible volume increase is zero and changes in Pc would be
proportional to changes in water and solute content of
that cell. Provided that changes in solute concentration
within the protoplast would be the only process that
generates a driving force for water uptake, a change in
turgor (APC) is a measure of the amount of solute
accumulated by this cell (/l/is), and APC is given by
(Steudle, 1989, 1992):
1
Cc+Cv,
(
X RT
{ ~V
A",
(3)
It can be seen from equation (3) that APC corresponds
to the increase in Anc (Anc = RTxAnJV) and AVUstuc.
A Vtiuuc denotes the net volume change of the entire cell
(protoplast and cell wall). Changes of nc do not directly
translate into APC. Rather, elastic and plastic deformation
of the cell (Fxe/(e + nc), e is the elastic coefficient of a
cell) and the storage capacities of the wall (Cw) and the
protoplast (Cc) dampen effects of An, on APC. The factor
F, which is approximately (1 + trjn xAt)'1 for <r»7rc,
accounts for the plastic extension (mAt) of a cell. In
contrast to elastic extension (1/e), plastic extension is
time-dependent; At denotes the time interval of relative
volume increase (AV/V). F approaches 1 for m->0 (i.e.
in a non-growing cell) and for At—>0. If water gain for
APC is entirely from the apoplast of the cell, AVtiuue = 0
and equation (3) becomes
e
Cw
C
c
+
w
An.
-T7V
(4)
Because of the smaller storage capacity of the wall
compared to the protoplast, the ratio CW/(CC + CW) is
significantly smaller than unity and changes of APC will
be small. This relationship may illustrate the situation for
shoot growth, when cell expansion strongly competes
with transpiration for water. Under these conditions,
Root elongation and drought
osmotic adjustment of the growing cell is rather inefficient
to maintain Pc > Y.
With sufficient water supply, i.e. water loss in the
wall due to uptake into the protoplast is immediately
replaced by an external water source, AVuauc equals
Cc x RTxAnJV(Steud\e,
1992). Thus, effects of compartmentation between apoplast and symplast cancel out and
equation (3) reduces to
APC =
An.
V
(5)
Here, An, translates almost entirely into turgor changes
provided that e»7r c . Equation (5) represents the corresponding turgor change to an increase in solute content
of an expanding cell in the root elongation zone when
water availability is unlimited. The net deposition rate of
solutes into the elongation zone would then be given by:
An,
At
APC
At
RT'
In the present study, AnjAt was calculated from previous
data of APJAt (Frensch and Hsiao, 1995). The calculations assumed F=\, because APJAt was measured at
zero root elongation and, hence, mAt = 0.
A different approach to obtain data on solute accumulation in tissues is to study growth in terms of fluid flow
(Silk, 1984). This model yields quantitative information
on the magnitude and the spatial distribution of growth
along the elongation zone of an organ. Accordingly, net
deposition rates of matter for a small root segment are
calculated from concentrations of the molecule of interest
within this segment and from the local growth rate of
this segment. For maize roots, deposition rates of water
and several major osmotic compounds have been determined for segments of 0.5 mm length (Sharp et al., 1990).
Local deposition rates calculated by this model may be
compared with measurements of AnJAt in single cells.
Mechanisms of adaptation to water deficits in
roots
It has long been recognized that the expansion rate along
the root elongation zone follows a bell-shaped curve. For
a maize root, the acceleration and deceleration phases
extent over a length of approximately 10 mm (Erickson
and Sax, 1956). More recently, elementary growth rates
have been combined with turgor measurements. These
studies show that turgor is rather uniform throughout
the cortex of the elongation zone in spite of large differences in local growth rates in wheat (Pritchard et al.,
1990, 1991) and maize (Spollen and Sharp, 1991).
Longitudinal gradients in wall rheology are therefore
likely the reason for the observed variability in local
growth rates (Tomos and Pritchard, 1994). Despite the
989
similarities in both species, a notable difference was
observed in these studies when roots were exposed to
water stress over medium long periods (hours to days).
While maize roots seemed to increase the wall yielding
ability, thus maintaining root growth at lower turgor,
wheat roots appeared to do just the opposite, i.e. they
reduced wall yielding ability. The discrepancy may indicate a fundamental difference between the species during
the adjustment to limited water supply.
Further investigations aimed to unravel primary
responses in the root elongation zone, including wall
rheology and water and solute transport properties, at
shorter time intervals (Frensch and Hsiao, 1994, 1995).
The results of turgor recordings during steady and
dynamic conditions of root growth can be summarized
as follows:
(i) Upon exposure of roots to hyperosmotic solutions, Pc
declined rapidly and growth decelerated or stopped almost
immediately. However, the inhibition of root growth was
compensated partially or completely within min due to
both a decline of the yield threshold (Y) and an osmotic
response GdwJ that led to a recovery of Pc. A minimum
pressure of 0.3 MPa was required for irreversible cell
expansion. During the short period of adjustment
(1-30 min) the wall extensibility coefficient (m) remained
rather constant.
(ii) The osmotic response of each cell was presumably
the result of carbohydrate uptake. Radial solute transport
in the elongation zone was sensitive to turgor changes.
With successively lower Pc in the cortex, the rate of solute
transport declined drastically.
(iii) Consistent with earlier results on root growth (Sharp
et al., 1988), the length of the elongation zone in maize
was reduced from originally 10 to approximately 7 mm
at i/(m< —0.5 MPa. The permanent inhibition of growth
in the proximal part of the elongation zone was likely
due to the lack of turgor recovery (solute transport).
Therefore, Pc remained smaller than 0.3 MPa, the minimum Y. Towards the apex growth resumed, because
radial solute flow continued and exceeded Y eventually,
(iv) The hydraulic conductivity of the tissue remained
high at all levels of Pc. Based on the hydraulic conductivity
coefficient of the tissue (K= 1.3 x 10"10 m2 s"'MPa"')
gradients of water potential across the cortex were calculated. At maximum expansion rates in the middle of the
elongation zone, gradients of only 0.03 MPa would be
expected.
On the premise that root elongation stops at Pc = Y,
and recalling that Pc was rather uniform across the cortex,
the yield coefficient of the tissue (y,) was determined
from the decline of turgor following osmotic step changes
in the medium (Fig. 3). The cross-hatched area indicates
the approximate pressure gradient in the cortex at which
the root elongation rate became zero. Because of the
small time interval (few seconds) it took for Pc to drop
990
Frensch
1
1
—
00 CO
Q_
'
1
•
A
2 , -0.2 -
—
5* -0.4 _
-
1
3 -
1
'
B
'.c
E
•§*-
-
2 -
CD
8 2
AAM.
1
•S
CO
A
La
_
1 -
O)
c
o
/
CD
/
0 _
0.7 - o
CO
i
c
i
•
dP D
grt3
i
a.
~^
'
n rW
A
/
throughout the cortex, and growth resumed eventually
once Pc exceeded the critical pressure. Thus, the yield
threshold of the tissue could be determined a second time
(Y2) from the kinetics of root elongation and turgor
pressure. For the experiment in Fig. 3, the yield threshold
declined approximately 0.15 MPa within 15 min, indicating rapid adjustment of the parameter to low Wm. As
soon as growth continued, turgor recovery leveled off and
reached a new steady state substantially smaller than the
original. Yield thresholds (either Yt or Y2) were by
approximately 0.1 MPa lower than the corresponding
steady turgor values, which implied that the mechanical
resistance (\/(m x V)) did not change significantly during
the treatment (equation 2).
After the removal of the osmoticum from the medium,
Pc rose and growth rates increased transiently to very
high values (Fig. 3B). Evidently, turgor increase generated a large driving force (Pc-Y2) f° r growth, which
presumably declined thereafter due to an increase of Y2.
Tur
8, 0.5-
1 j^zv____, p ^
The variable nature of the yield threshold
A major point of the kinetic study is the observation of
the non-linear relationship between root elongation and
I
turgor. Largely, the non-linearity is caused by the variable
11
12
13
14
nature of Y. Within 2-30 min, Y adjusted to a new value
Time (h)
ranging between 0.7 and 0.3 MPa for positions 3 to
Fig. 3. (A) Typical response curves of root elongation rate and turgor
10mm behind the root tip (Frensch and Hsiao, 1995).
pressure (/>c) to step changes of water potential in the medium (V.J.
Apparently, 0.3 MPa was the minimum turgor required
(B) Root elongation rate was determined by a position transducer as
for growth. The range represents the 'instantaneous'
shown in Fig. 1. In (C), turgor response was measured in cortical cells
at different radial positions (distance to root periphery: V, 20-40 ^m;
ability of the root to compensate for turgor changes.
O, 50-90fxm; —, 50-90^m; D, 100-160 ^m). The solid lines in (C)
Changes of Y are likely to correspond with metabolic
indicate periods of continuous measurements during pressure relaxaprocesses in the cell wall which trigger both the weakening
tions; the symbols indicate measurements of short intervals. Upon the
addition of the osmoticum, a biphasic pressure response was observed.
and hardening of the mechanical network, probably by
The first phase was due to water loss to the medium and was used to
altering the number of load-bearing tethers between the
determine the yield threshold of the tissue (K,) (elongation rate = 0) as
microfibrils (Passioura and Fry, 1992).
well as the hydraulic conductivity coefficient. While elongation was
zero, turgor recovered during the second phase and reached a new
The dynamics of changes in Y are perhaps best demonsteady level at 0.5 MPa. The amount of turgor recovery due to solute
strated
by continuous recordings of pressure relaxations
accumulation was almost 0.2 MPa Elongation resumed when Pc
when
water
was withheld from entering the tissue
reached the yield threshold (Y2), which was already significantly lower
than K,. After the withdrawal of the osmoticum, elongation rate
(Cosgrove, 1987, 1993). In these experiments, turgor
increased transiently to high values. This increase was associated with
decayed in a nearly exponential fashion to new steady
a large growth-effective turgor (Pc-Y). The shaded areas indicate
levels, ranging between 0.1 and 0.3 MPa for various
possible turgor gradients across the tissue during the determinations of
K, and Y2- The dashed line indicates the possible kinetic of the variable species. Assuming that Y remained constant throughout
Y. Growth-effective turgor were calculated from the differences (Pc-Yt)
the pressure relaxation, Cosgrove concluded that the final
and (Pc-K2). (Adapted from Frensch and Hsiao, 1994.)
pressure would indicate the yield threshold of the corresponding tissue. However, in light of the new results on
the variability of Y, the basis of interpretation could
from the initial level to Yu this value probably represents
be quite different. Hence, the pressure relaxation in
the steady state conditions in the root before the
osmoticum was added. While root growth was zero, PQ Cosgrove's experiments may represent the adjustment of
Y to the water stress treatment. Accordingly, the final
further declined to a minimum. If the cortical cells in the
pressure
would correspond to the minimum Y. This
elongation zone would behave like simple osmometers
interpretation
is in accordance with measurements on
with semipermeable membranes, as found for mature cells
growing
soybean
stems using the isopiestic psychrometer
for the short-term duration of the experiment (Frensch
technique
(Boyer
et
al., 1985). The authors detected rapid
and Hsiao, 1994), the pressure would be expected to
pressure
relaxations
of approximately 0.1 MPa to the
remain at the low level. However, turgor recovered
0.3 -
turgor recovery
Roof elongation and drought 991
presumed Y, followed by a slower pressure reduction
afterwards.
Consistent with the results on leaf growth and results
of other studies, (Pc-Y) for roots was small and cell
expansion responded to slight changes in Pc (Matyssek
et al., 1988; Okamoto et al, 1989). Compared with the
conventional method, i.e. when Y is calculated from
steady state relationships between growth rate and Pc,
the direct approach yielded much higher values of Y and,
thus, smaller values of (Pc-Y) (Frensch and Hsiao, 1994).
The discrepancy may caution to evaluate Y and m applying the linear relationship of the Lockhart equation to
experiments where the water status of the root is altered.
Earlier conclusions drawn on the assumption that Y
behaves like a constant may be premature.
3 -
I
i
i
A
2 CO
O)
root
1 -
leaf
LLJ
0 -
Osmotic
experiment
i
30
Nutrient solution withheld
i
i
60
90
-
120
Time (min)
Is the determination of Y confounded by elastic
responses?
In principle, elastic properties of cells and tissues affect
the time-course of pressure responses. For example, substantial instantaneous and retarded elasticity was reported
for excised coleoptiles of maize using extensiometer techniques (Hohl and Schopfer, 1992). In the experiments by
Frensch and Hsiao, the direct determination of Yx and
Y2 in roots was particularly critical, because longitudinal
shrinkage of the entire root would contribute to the
growth recording using an LVDT. Frensch and Hsiao
(1995) excluded a significant contribution of elastic longitudinal shrinkage for maize roots on the basis of experimental results on mature root tissue. Further evidence in
favour of the correct determination of Y is derived from
additional experiments on plants which include the
elongation zone (Fig. 4). It can be seen that osmotic
pressure steps in the absence of growth (Pc < Y) did
neither result in axial elastic shrinkage or swelling of the
root. Radial elastic responses, which are likely to occur,
would not contribute to the determination of Y.
Besides mature xylem acting as a 'backbone' in the
root, the experiments provide new evidence that a low
water status of the tissue adds to the elastic response of
the root. Shrinkage of the root could be induced during
the later part of the experiment, when the nutrient solution
was completely withheld from the root and Pc presumably
dropped to zero (Fig. 4). Thus, elastic coefficients of
single cortical cells in the order of a few MPa (Pritchard
et al., 1990; Zhu and Steudle, 1991) may be useful to
calculate radial elastic responses of roots, but are inadequate for longitudinal estimations. This relationship is
extremely important for sustained uptake of water and
nutrients. Significant changes in root length due to
changes in water potentials in the xylem or soil would be
expected to disrupt the contact between root and soil,
established by root hairs and mycorrhizae. It is yet to be
determined whether irreversible root extension in the
basal part of the elongation zone is limited by the
25
Time (min)
Fig. 4. Test of longitudinal elastic properties of a maize plant mounted
to a setup shown in Fig. 1. In (A), the time courses of root and leaf
elongation are shown in response to induced water deficits in the
medium. In (B) and (C), the osmotic experiment is shown at higher
resolution. As expected from previous experiments, no further elongation was recorded by the root position transducer following a decrease
of the water potential in the medium to - 0 . 2 MPa using mannitol. The
pressure decline reduced turgor pressure below the yield threshold
(PC<Y) and elongation stopped. The plastic (irreversible) response
could have been masked by elastic responses of the root, which were
canceled out during the following osmotic pressure steps. In the absence
of any elongation, additional osmotic step changes had no effect on
longitudinal shrinkage and swelling in the root and leaf. Root shrinkage
could be induced, however, after the water supply was cut off
completely.
mechanics of progressively mature protoxylem in this
zone. This proposition would differ from that postulated
for stems and other aboveground organs, where growth
appears to be limited by the expansion rate of the outer
epidermis (Kutschera, 1995).
The role of solute accumulation for adjustment of
growth to water deficits
The pattern of local growth rates along the elongation
zone is associated with different rates of matter (water
and solute) deposition (Sharp et al., 1988, 1990).
Moreover, the adjustment of local root growth to moderate water stress (4>m> — 0.6 MPa) correlates with the
ability of the root to maintain continuous solute uptake
992
Frensch
(Pritchard et al., 1991; Frensch and Hsiao, 1994, 1995).
This is well illustrated by the different rates of turgor
recoveries with respect to the amount of water deficits
and to the position along the expansion zone (Fig. 5). At
distances 4-5 mm behind the apex, dP/dt decreased by
approximately one order of magnitude with increasing
water stress (range of *Fm: —0.1 to —0.6 MPa). It appears,
therefore, that solute transport in the elongation zone is
turgor-sensitive, as has been shown previously for carbohydrate transport in other tissues (Oparka and Wright,
1988; Patrick, 1994). At f m < - 0 . 5 MPa, the turgor
recovery in maize roots was confined to positions < 7 mm
(Frensch and Hsiao, 1995). The results imply that the
shortening of the growing zone, which was also found by
others (Sharp et al., 1988; Pritchard et al., 1993), was
caused by the lack of solute accumulation to raise Pc
above the minimum Y of 0.3 MPa.
Evidence for the phloem as the major source for solute
accumulation is coming from different slopes of the turgor
recovery at various positions in the cortex and from
transient turgor gradients across the cortex at zero growth
rates (Frensch and Hsiao, 1994). In these studies, the
maximum rate of turgor recovery (AP/At) increased substantially from cells at the root periphery toward the
endodermis. These experiments were carried out at zero
growth rates, i.e. when dilution effects due to cell enlargement did not contribute to the effect.
Maximum rates of turgor recovery may be used to
estimate the net deposition rates of osmotically active
compounds into the elongation zone according to equation (6). For positions 4-10 mm behind the apex, an
average AP/At of 3 x 10~4 MPa s"1 was evaluated for
the cortex following a step change in x¥m from zero to
—0.3 MPa (Frensch and Hsiao, 1995). Using measured
root dimensions (radius of root=400 ^m, radius of stele =
150 ^m) and assuming a value of 0.8 for the term e/(e + 77c)
(Frensch and Hsiao, 1995), the net deposition rate into
the cortex for an increment of 1 mm, Anj'At = 235
nosmolmm"1 length h" 1 at 7=298 K. Accordingly,
AnJAt = l&5 and 118 nosmolmm"1 length h" 1 for ¥/m =
— 0.1 MPa and —0.6 MPa, respectively. Compared to
maximum net deposition rates based on the evaluation
of growth kinematics (Sharp et al., 1990), the data by
Frensch and Hsiao were larger by factors of 2 to 5. It
would be premature, however, to analyse the differences
quantitatively from the present set of data. Nevertheless,
they indicate that processes other than solute uptake may
have contributed to the increase of Pc as well.
The coupling of solute transport and turgor adjustment
in growing tissues has been investigated in hypocotyls of
Ricinus (Meshcheryakov et al, 1992). Unlike roots, large
radial turgor gradients were found which corresponded
to a gradient in osmotic pressure of the same amount.
Thus, the sum of both, the water potential, was essentially
homogenous throughout the tissue. The results imply that
solute transport from the center to the periphery is slow
compared to water transport, and additional perfusion
experiments supported the theoretical prediction that the
regulation of the water potential in the symplast was very
sensitive to small changes of the solute content in the
apoplast (Steudle, 1992). The compartmentation of solutes between symplast and apoplast may also account for
water potential gradients across the tissue, which are
usually associated with the process of wall loosening
(Cosgrove, 1993; Nonami and Boyer, 1993). More work
is needed to understand the role of compartmentation
Distance behind root tip:
+ 0.11 MPa
1.0 -r
•
— o
0.8 --
3 - 4 mm
4 - 5 mm
6 - 7 mm
7 - 8 mm
CO
a.
06 --
O
O) 04 -0 2 -•
0.0 -120
40
60
Time (min)
Fig. 5. Three pressure kinetics of cortical cells in the root elongation zone following osmotic step changes in the medium (KCl, mannitol). With
increasing water deficits, turgor pressure (Pc) declined rapidly from initial values (range 0.6-0.7 MPa) to a minimum pressure Subsequently, Pc
recovered completely (water potential in medium, fm: —0.11 MPa) or partially (>?„: -0.29 and -0.57 MPa) presumably due to the uptake of
solutes at zero growth rates. However, the maximum rate of turgor recovery (dP/dt) differed substantially and declined by about one order of
magnitude at higher water deficits. Moreover, the measurements in different parts of the elongation zone indicate that turgor recovery stopped
completely at more basal positions (7 mm and more behind the tip), Due to the loss of turgor adjustment, Pc remained smaller at those positions
than the minimum yield threshold for growth, 0.3 MPa.
Roof elongation and drought
and, especially, that of the apoplast within the complexity
of tissue growth (Steudle and Frensch, 1996).
993
Unstressed root
What limits root growth?
As already stated, limitations of growth are usually
discussed in terms of hydraulic and mechanical resistances. The relative magnitude of the resistances can be
estimated from the ratio (Boyer, 1985)
LpA
Pc-Y
(7)
according to the Lockhart theory. Values significantly
smaller than 1 indicate a growth-limitation by the wall
extensibility. Using measured values of (Pc — Y) and calculated water potential gradients in growing maize roots
(Silk and Wagner, 1980; Frensch and Hsiao, 1995), the
ratio of mechanical versus hydraulic resistance ranges
between 0.1 and 0.5 for unstressed and moderately
stressed roots, depending on the axial and radial position
of the expanding cell in the tissue. Cells at their maximum
local growth rate, which make up most of the over-all
growth rate, have values closer to 0.5, which suggests
that both (or neither) of the parameters is rate-limiting.
It is tempting to argue that solute accumulation could
have become the rate-limiting process for growth under
water stress. In favor of this hypothesis are the possible
pressure-sensitive distribution of carbohydrates and the
close relationship between the reduced length of the
elongation zone and the loss of turgor recovery (Frensch
and Hsiao, 1995). Carbohydrates account for approximately 50% of the net solute deposition rate in maize
roots (Sharp et ai, 1990). They arrive from the shoot
and are distributed radially into the growth zone along
the symplastic route (Fig. 6). The flow rate of sugars
decreases progressively with farther distance from the
phloem until it eventually drops to zero in the epidermis.
Simultaneously, water enters the tissue largely across the
epidermis, driven by water potential gradients induced by
either wall relaxation or solute gradients along the radial
path. Under steady conditions, a 'functional equilibrium'
is expected for this position, where the local growth rate
is determined by cells limited in either carbohydrate or
water supply. The ratio between the rate constants of
water and solute flow would then determine whether the
overall growth rate is closer associated with cell layers at
the periphery or with those in the center of the root.
With increasing water stress and lower levels of Pc, the
results in Fig. 5 suggest that the epidermis could become
the growth-limiting cell layer in the root, provided that
(Pc-Y) can adjust to positive values. On the other hand?
turgor recovered rapidly in cortical cells following small
perturbations in the medium ( f m > = —0.1 MPa). This
indicates that there is no obvious limiting factor or tissue
during steady growth conditions.
Carbohydrate
flow
»
Water flow
Low medium water potential
Fig. 6. Schematical drawing of a cross section of the maize root
illustrating radial water and carbohydrate flows in the elongation zone
for unstressed and water stressed roots. The location of the phloem is
represented by the cross-hatched ring. The arrows indicate the directions
of flow, and the thickness approximates the radial flow rates within the
growing tissue.
There is growing evidence that plasmodesmata represent the major resistance for symplastic solute flow (BretHarte and Silk, 1994). Dye-injection studies suggest that
symplastic unloading from the phloem ceases when root
growth is reduced (Oparka et al., 1995). Although the
reduced permeability would explain the results presented
in the present study, there is also evidence for the contrary.
Using electron microscopy, Schulz (1995) reports a
widening of plasmodesmata in pea roots in response to
osmotic step changes in the medium, which would
enhance symplastic solute flow at constant concentration
gradients across the tissue. Perhaps, the lack of turgor
recovery correlates with a smaller driving force for phloem
unloading at low fm, which requires further attention in
future experiments. In either case, solute transport into
the elongation zone appears critical for continuous root
growth in water-stressed and unstressed roots.
Growth limitations: scaling up from cell to
organ level
So far, the discussion on interactions between cell
enlargement and water shortage has been confined to the
994
Frensch
elongation zones of the root and leaf. Different plant
organs, however, do not experience the same amount of
water stress which relates to the observed differential
growth responses of the two organs (Fig. 1). The gradients in water potential between the elongation zones in
root and shoot largely depend on the hydraulics of the
root system. Therefore, growth limitations as related to
aspects of water transport in roots shall now be discussed.
Water loss of the plant to the atmosphere generates a
tension in the xylem which, according to the cohesion
theory, propagates into the root and reduces the water
potential of the xylem below that of the soil. Limited by
hydraulic resistances in series and parallel, water moves
along the gradient from the soil to the shoot. It is useful,
therefore, to separate the volume flow into a radial and
an axial component. The longitudinal flow is solely driven
by hydrostatic pressure gradients in the xylem. Within
the radial pathway, water potential gradients may be
hydrostatic, osmotic, or matric in nature. In the absence
of transpiration, the osmotic gradient dominates, which
leads to guttation and xylem exudation in many plants.
Radial water flow in roots is commonly analysed
applying a two-compartment model in which the xylem
is separated from the soil by a complex barrier (Passioura,
1988; Steudle, 1994; Steudle and Frensch, 1996). The
application of the root pressure probe technique to roots
of herbaceous and woody species has yielded the remarkable result that the radial hydraulic conductivity of the
root (Lpr) strongly depended on the nature of the driving
force. In tree roots, for example, Lpr(hydrostatic) was
larger than L/;r(osmotic) by one to three orders of magnitude, whereas no difference between L/?r(hydrostatic) and
L/?r(osmotic) was found in Phaseolus and barley (Steudle,
1994). Radial volume flow rates in maize are larger by
approximately one order of magnitude in the presence of
hydrostatic pressure gradients compared to osmotic
(Fig. 7). The experiments on plants of different age
show that discrepancies between LpT(hydrostatic) and
L/>r(osmotic) were found consistently in young roots as
well as in three-weeks old root systems, when the length
of the main root axis exceeded 600 mm.
The data in Fig. 7 further illustrate a decreasing efficiency of roots in water uptake when the main axis exceeded
a length of approximately 200 mm. This result is not
surprising if one considers the developmental gradients
along the root axis. From the apex toward the base, the
function of the root shifts from a tissue more specialized
in material uptake to one more specialized in material
translocation. To some extent this process may be compensated by the development of laterals, although this
effect appeared to be small in the present experiments on
maize roots. For roots exposed to uniform environmental
conditions, water uptake characteristics in hydrostatic
experiments remained rather constant along the apical
150 mm. The decline of L/?r(hydrostatic) by one order of
i
,-'nj
30 -
,r
20 -
\n
1
1
1
o
•
o
o
1
Lp, (hydrostatic) _
Lp, (osmotic)
-
o
E
o
10 0 1
0
200
1
1
400
1
I
600
Length of main root axis (mm)
Fig. 7. Radial hydraulic conductivity (Lp,) of maize roots (age 2-24 d)
grown in solution culture The primary root of a maize plant was
excised and attached to the root pressure probe (Steudle and Frensch,
1989). Hydrostatic pressure steps were induced by means of the probe;
osmotic step changes were applied by replacing the root medium with
a solution of higher osmotic pressure. Both experiments resulted
in a radial flow of water which caused a pressure relaxation in the
xylem. The two types of pressure relaxation experiments yielded
L/>r(hydrostatic) and L/>r(osmotic). Each data point is the mean of
2-14 measurements on a single root. The data on maize roots show
two important results: L/> r (hydrostatic)>Lp r (osmotic), and Lp, declines
with increasing root size The dashed lines indicate Lp,(hydrostatic)
before the emergence of laterals (25 x 10~ 8 m s" 1 MPa" 1 ) and after
laterals amounted for more than 90% of the total root surface area
(2.2x IO" 8 m s"
magnitude coincided with the growth of root laterals,
emerging from the cortex approximately 150 mm behind
the apex. After 14 d in nutrient solution, when the main
axis had reached lengths of approximately 500 mm,
laterals accounted for more than 90% of the total root
surface area (Frensch, 1990). Hence, small LpT of older
roots largely represents the hydraulic conductivity of the
laterals. It may be concluded that the laterals of maize
roots grown in solution culture are less efficient in terms
of water uptake and transport than the main axis.
Radial water transport in roots is composite in nature
(Steudle, 1994). The corresponding model involves two
parallel pathways: an apoplastic and a cell-to-cell path,
each exhibiting different hydraulic and osmotic properties.
It has been supported by pressure clamp experiments on
Prunus using a different experimental and theoretical
approach (Magnani et a!., 1996). Despite its advantages,
it is worth noting that this model simplifies the process
of water transport in roots considerably. Root development and function depends much on the physical and
chemical conditions of the immediate environment, which
is largely unexplored for most plants under natural conditions (Waisel et al., 1991; McCully, 1995). Both axial
and radial resistances as well as water potential gradients
vary with time and space along the root. The implications
for the hydraulic architecture of the root have been
studied theoretically and experimentally on excised and
intact roots using pressure probe techniques (Frensch and
Roof elongation and drought
Steudle, 1989; Frensch and Hsiao, 1993; Frensch et al.,
1996), and will be subject of the following chapter.
Pressure propagation across the root and
implications for leaf and root growth
Information on soil water content is transmitted to the
shoot by hydraulic and metabolic 'signals' originating in
the root. They modulate stomatal movement and shoot
growth within complex control loops (Tardieu and
Davies, 1993). Over the past decade, the action of the
phytohormone abscisic acid has been investigated extensively, whereas the meaning of the hydraulic component
was largely neglected. To some extent, the negligence may
result from the lack of appropriate techniques to measure
and to modify xylem pressures in plants. Numerous
measurements of xylem water potential by indirect
methods (e.g. psychrometry, Scholander's pressurechamber technique) have predicted substantial tensions
in the xylem of a transpiring plant. Recent studies have
yielded satisfactory results on the reliability of the pressure
chamber technique, i.e. to correlate applied (positive)
pressures with tensions in excised plant material
(Holbrook et al., 1995; Pockman et al., 1995). Data of
negative pressures in these studies are in accordance with
the cohesion theory, which has prevailed the literature on
water movement in plants over the last 100 years.
However, the dogmas of this theory have been attacked
on the basis of direct measurements in the xylem using a
modified cell pressure probe (Balling and Zimmermann,
1990; Zimmermann et al., 1993). The authors report
higher (less negative) values in the xylem of herbs and
tree species compared to Px obtained by the pressure
chamber technique. Direct xylem measurements in trees
indicated that longitudinal pressure gradients were too
small to move water from the roots to the leaf by a purely
hydraulic mechanism (Benkert et al., 1995). If true, the
results of Zimmermann and co-workers would change
basic concepts of the mechanism of long-distance transport in the xylem. However, the results have to be taken
cautiously. Due to technical limitations of the cell pressure
probe in the negative pressure range, which have not been
addressed satisfactorily in the studies by Zimmermann
and co-workers, the critique against the cohesion theory
has been refused (Steudle, 1995).
The existence of longitudinal pressure gradients is a
key feature of the cohesion theory and also relevant for
plant growth. How effectively do pressure changes in the
shoot propagate into the root and further into the soil?
Recently, the propagation of hydrostatic pressures was
measured in the cortex and in the xylem of maize root
systems (Frensch and Hsiao, 1993; Frensch et al., 1996).
These experiments provided novel results on the interactions between root structure and function. When the
shoot was replaced by a modified root pressure probe,
995
volume flows could be induced either into or out of the
cut end of the excised root (Fig. 8). The cell pressure
probe was used to scan the kinetics of the pressure
propagation along mature and immature xylem (AP^) of
root systems up to 500 mm in length. As expected, APipp
was rapidly transmitted (fractions of a second) along
conductive xylem. Longitudinal pressure propagation was
determined from constant values of APX and APhpp.
Ratios of APJAPBpp ranged from values close to 1 at the
base of the root to almost zero at the root tip (Frensch
et al., 1996). Based on measured APJAPnpp, a hypothetical pressure profile was calculated for the boundary
condition Px= - 0 . 5 MPa atz = 500 mm and Wm = 0 MPa
(closed symbols in Fig. 8). The data show that a tension
in the shoot is maintained for a long distance in the root
xylem, but steeply attenuates toward zero in the apical
200 mm.
The attenuation of xylem pressure is important for the
differential growth of roots and shoots in the presence of
water deficits in the atmosphere and in the soil. In our
example, the leaf elongation zone would experience a
tension in the xylem (apoplast) of —0.5 MPa or less,
while Px in the root elongation zone would be almost
zero. Frensch and Hsiao (1993) concluded that root
elongation was rather hydraulically isolated from changes
in water potential that originate in the shoot. On the
contrary, water potential gradients along the root are
different when water stress is imposed through osmotic
changes in the medium. In the absence of significant
transpiration, the osmoticum reduces P% more evenly over
the entire length of the root. Due to reflection coefficients
of the root smaller than unity, values of APX are smaller
than the corresponding A-nm (Steudle and Frensch, 1989).
Hence, root growth is probably more affected than shoot
growth when Wm is altered.
Assuming that transpiration is the major driving force
for water transport, the experiments on pressure propagation along the xylem resemble the situation in the intact
plant. Further steps toward an understanding of dynamic
pressure changes in the xylem under transpiring conditions require direct measurements in the xylem. These
experiments were performed on plants kept in either
solution culture or in soil (Frensch, Dieffenbach and
Gottlein, unpublished results). Reliable tests of the cell
pressure probe along with further control experiments
demonstrated convincingly that the measurements were
performed in conductive xylem. Negative xylem pressures
of down to —0.5 MPa absolute were recorded in roots
of maize plants. They depended not only on transpirational water loss, but also on the availability of water
in the medium and soil. Essentially, the results verified
that transpiration generates a rapid and effective driving
force for water uptake in the xylem. Furthermore, the
studies explain the apparent lack between transpiration
and xylem pressure, which was used by Zimmermann and
996
Frensch
Longitudinal pressure propagation
pressure transducer
500
rrdcroeyringe
modified root
pressure probe
pneumatic
pressure —
8lllcone oil
I
*
z = 150 mm
longitudinal
pressure
propagation
radial pressure
propagation
cell pressure
probe
z = 50 mm
I
z « 0 mm
oot elongation zone
-0.6
-0.4
-0.2
0.0
Xylem pressure (MPa)
Fig. 8. Experimental set-up and results of longitudinal pressure propagation on excised roots. With the syringe detached, the root pressure probe
recorded steady root pressures of maize roots. Once the syringe was introduced, the pressure in the probe was clamped to a smaller or higher level
than root pressure by means of compressed air. The arrows illustrate the directions of water flow following a pressure application higher than root
pressure (APtvp); the thickness of the arrows indicate that radial and axial flow varies along the main axis. The direction of flow could be reversed
when P,m in the probe was clamped to a value below the original root pressure. The bars to the left of the root axis indicate the zones of xylem
maturation (PX = protoxylem; EMX = early metaxylem; LMX = late metaxylem) which affected the longitudinal pressure profile shown to the right.
Throughout the root, pressure response (APJ in mature (conductive) EMX and immature LMX was monitored with the cell pressure probe.
Measured ratios of APJAPtpp were multiplied by —0.5 MPa (closed symbols) to illustrate a hypothetical steady profile for a transpiring plant The
profile indicates that pressure attenuation toward the tip 'isolates' the root elongation zone hydraulically from changes in xylem tensions generated
in the shoot. The solid line was modeled using measured hydraulic resistances for axial and radial water transport determined by the root pressure
probe technique according to the cable theory (see text). (Adapted from Frensch, 1990 and Frensch et al., 1996.)
co-workers to dismiss the cohesion theory, within the
concept of the soil-plant-atmosphere continuum.
In summary, when restricted to purely hydraulic
aspects, the results indicate that root elongation is strictly
coupled to the water potential of the immediate external
surrounding. Shoot growth, on the other hand, is closely
linked to changes in air humidity, provided that significant
changes in Px are induced by transpiration. If the phJoem
is included in these considerations, root elongation may
also respond immediately to hydraulic changes in the
shoot. It is common to consider that carbohydrate flow
into the sink region depends on a hydrostatic pressure
gradient along the phloem, which would be diminished
by a rapid decline of water potential in the shoot, at least
transiently. As discussed before, carbohydrate supply
appears to be a critical factor for the rate of root
elongation, especially when turgor in expanding cells
is low.
Modelling of water potential gradients along the
xylem
What caused the observed pressure attenuation in the
root? This leads to a two-dimensional mathematical
model of water flow in roots, adapted from a model of
electric current flow along a leaky cable ('cable theory';
Taylor, 1963). The spread of nerve impulses along an
axon is a more prominent example of the application of
this theory (Rail, 1987). It extends the two-compartment
(one-dimensional) analysis of water uptake introduced
before by a longitudinal flow in the xylem. Applied to
roots, the cable theory predicts gradients of longitudinal
xylem pressure on the basis of two parameters: radial
(RR) and axial resistances (Rx)- Assuming constant values
of 7?R and Rx, the equations of the cable theory can be
solved analytically (Landsberg and Fowkes, 1978). To
model pressure profiles and cumulative flow rates in roots
more realistically, the variable nature of hydraulic
Root elongation and drought
resistances was incorporated in the theory (Frensch and
Steudle, 1989; Frensch, 1990). Using the root pressure
probe technique to evaluate RR and Rx (Frensch and
Steudle, 1989; Frensch, 1990), the pressure attenuation
along the xylem was modeled for the maize root system
(Fig. 8, solid line). Similar pressure profiles were calculated for roots of Agave and Opuntia (Aim et al., 1992).
It can be seen in Fig. 8 that the two independently
determined profiles correspond reasonably well. Discrepancies between the profiles are probably due to the
fact that the model (i) neglects the distribution of resistances within the laterals (largely unknown), and (ii)
overestimates the contribution of axial resistances in the
tip region (0-20 mm), which is difficult to assess with the
root pressure probe.
According to the cable theory, the ratio RfJRx is a
fundamental parameter for the longitudinal pressure
gradient in the xylem and, hence, for the distribution of
water potential differences between the xylem and the
root surface. An increase of Rx would be associated with
a higher proportion of immature xylem. Because resistances decrease with the forth power of the vessel radius,
changes of Rx due to xylem maturation are huge. With
the differentiation of the early and late metaxylem, ^?x
decreased by 5 orders of magnitude in maize roots
(Frensch, 1990); changes in Rx by 3 orders of magnitude
were reported for onion roots (Melchior and Steudle,
1993). Both studies emphasize to measure Rx rather than
to estimate this parameter from Poiseuille's law because
of significant differences between the two approaches.
Changes in RR along the main root axis are affected by
changes in both the apoplastic and cell-to-cell path. In
older parts of grass roots, the endodermis is a likely
candidate for the radial hydraulic barrier (Sanderson,
1983; McCully and Canny, 1988). The dominating role
of the endodermis for radial as well as axial pressure
attenuation was demonstrated for maize roots (Frensch
et al., 1996). The results are consistent with measurements
of water uptake in various root zones of barley and
spruce (Sanderson, 1983; Haussling et al., 1988). Within
the radial pathway, the lateral walls of mature vessels
may also pose a significant resistance to water uptake
(Peterson and Steudle, 1993), whereas the contribution
of developmental changes in the hypodermis appears to
be rather small (Clarkson et al., 1987).
Taken together, experimental data on pressure propagation in the xylem and cortex validate predictions of the
cable theory. The model provides a sound basis to study
the relationship between root structure and function at a
mechanistic level. Future work should aim to include root
growth and the responses of growth to water stress as well.
Conclusions
It has been known for a long time that cell expansion is
extremely sensitive to changes in plant water potential,
997
and that growth recovers rapidly from mild and moderate
water stress. Recent progress toward a better understanding of the underlying processes is highlighted by this
review. Measurements ranging from the cell to the organ
level emphasize to study growth responses of plants at
both high spatial and temporal resolution. The results
confirm the fundamental relationship between turgor and
cell expansion, which is usually masked when conventional approaches are applied, i.e. when steady conditions
of turgor and growth are compared. In the past, this has
led to confusion about the meaning of turgor for cell
enlargement.
Evidence is presented that rapid modifications of the
yield threshold as well as turgor-dependent solute transport into the growth zone are important features of roots
during the process of adjustment. From these data it
should be clear that growth and its recovery from transient inhibition is closely linked to solute transport. An
adequate model of tissue expansion, which is still missing,
should include pathways, mechanisms, and amounts of
matter supplied by different compartments.
The differential responses of root and shoot growth are
related to longitudinal water potential gradients in the
plant. The attenuation of xylem pressure in roots is
readily explained by the 'cable theory'. It combines features of root structure (hydraulic resistances) and root
function (water transport) and demonstrates that the
elongation zones in the root and shoot of maize are
hydraulically separated, consistent with previous experimental results. The data emphasize the need to study
interactions between growth and water relations on a
whole plant level.
Acknowledgements
The research was supported by the Award of a Feodor LynenFellowship from the Alexander von Humboldt-Foundation
(Germany) and by a Fellowship from the Deutsche
Forschungsgemeinschaft to JF. I gratefully acknowledge the
support of Drs Theodore C Hsiao, University of California,
Davis, and Ernst Steudle, University of Bayreuth.
References
Acevedo E, Hsiao TC, Henderson DW. 1971. Immediate and
subsequent growth responses of maize leaves to changes in
water status. Plant Physiology 48, 631-6.
Aim DM, Cavelier J, Nobel PS. 1992. A finite-element model
of radial and axial conductivities for individual roots:
development and validation for two desert succulents. Annals
of Botany 69, 87-92.
Balling A, Zinunennann U. 1990. Comparative measurements
of the xylem pressure of Nicotiana plants by means of the
pressure bomb and probe. Planta 182, 325-45.
Benkert R, Zhu JJ, Zimmennann G, TOrk R, Bentrup F-W,
Zinunennann U. 1995. Long-term xylem pressure measurements in the liana Tetrastigma voinierianum by means of the
xylem pressure probe. Planta 196, 804-13.
998
Frensch
Boyer JS. 1985. Water transport. Annual Review of Plan!
Physiology 36, 473-516.
Boyer JS, Cavalieri AJ, Schulze E-D. 1985. Control of the rate
of cell enlargement: excision, wall relaxation, and growthinduced water potentials. Planta 163, 527-43.
Bradford KJ, Hsiao TC. 1982. Physiological responses to
moderate water stress. In: Lange OL, Nobel PS, Osmond
CB, Ziegler H, eds. Encyclopedia of plant physiology Vol. 12B.
Physiological Plant Ecology If. Water relations and carbon
assimilation. Berlin: Springer-Verlag, 263-324.
Bret-Harte MS, Silk WK. 1994. Nonvascular, symplasmic
diffusion of sucrose cannot satisfy the carbon demands of
growth in the primary root tip of Zea mays L. Plant
Physiology 105, 19-33.
Carpita NC, Gibeaut DM. 1993. Structural models of primary
cell walls in flowering plants: consistency of molecular
structure with the physical properties of the walls during
growth. The Plant Journal 3, 1-30.
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.
Cosgrove DJ. 1987. Wall relaxations in growing stems: comparison of four species and assessment of measurement techniques. Planta 171, 266-78.
Cosgrove DJ. 1993. Wall extensibility: its nature, measurement
and relationship to plant cell growth. New Phvtologist
124, 1-23.
Erickson RO, Sax KB. 1956. Elemental growth rate of the
primary root of Zea mays. Proceedings of the American
Philosophical Society 100, 487-98.
Frenscb J. 1990. Wasser- und Teilchentransport in Maiswurzeln.
PhD thesis, Bayreuth, Germany.
Frensch J, Hsiao TC. 1993. Hydraulic propagation of pressure
along immature and mature xylem vessels of roots of Zea
mays measured by pressure probe techniques. Planta 190,
263-70.
Frensch J, Hsiao TC. 1994. Transient responses of cell turgor
and growth of maize roots as affected by changes in water
potential. Plant Physiology 104, 247-54.
Frensch J, Hsiao TC. 1995. Rapid response of the yield
threshold and turgor regulation during adjustment of root
growth to water stress in Zea mays. Plant Physiology
108, 303-12.
Frensch J, Hsiao TC, Steudle E. 1996. Changes of water and
solute transport along developing maize roots. Planta 198,
348-55.
Frensch J, Steudle E. 1989. Axial and radial hydraulic resistance
to roots of maize. Plant Physiology 91, 719-26.
Green PB, Erikson RO, Buggy J. 1971. Metabolic and physical
control of cell elongation rate—in vivo studies in Nitella.
Plant Physiology 47, 423-30.
HSussling M, Joins CA, Lehmbecker G, Hecht-Buchholz C,
Marschner H. 1988. Ion and water uptake in relation to root
development in Norway spruce (Picea abies (L.) Karst.).
Journal of Plant Physiology 133, 486-91.
Hohl M, Schopfer P. 1992. Physical extensibility of maize
coleoptile cell walls: apparent plastic extensibility is due to
elastic hysteresis. Planta 92, 498-504.
Holbrook NM, Burns MJ, Field CB. 1995. Negative xylem
pressure in plants: a test of the balancing pressure technique.
Science 270, 1193-4.
Hsiao TC. 1973. Plant responses to water stress. Annual Review
of Plant Physiology 24, 519-70.
Hsiao TC, Jing J. 1987. Leaf and root expansive growth in
response to water deficits. In: Cosgrove DJ, Knievel DP, eds.
Physiology of cell expansion during plant growth. The
American Society of Plant Physiologists, 180-92.
Kramer PJ, Boyer JS. 1995. Water relations of plants and soils.
San Diego: Academic Press.
Kutscbera U. 1995. Tissue pressure and cell turgor in axial plant
organs: Implications for the organismal theory of multicellularity. Journal of Plant Physiology 146, 126-32.
Landsberg JJ, Fowkes ND. 1978. Water movement through
plant roots. Annals of Botany 42, 493-508.
Lockhart JA. 1965. An analysis of irreversible plant cell
elongation. Journal of Theoretical Biology 8, 264-76.
Magnani F, Centritto M, Grace J. 1996. Measurement of
apoplasmic and cell-to-cell components of root hydraulic
conductance by a pressure-clamp technique. Planta 199,
296-306.
Matsuda K, Riazi A. 1981. Stress-induced osmotic adjustment
in growing regions of barley leaves. Plant Physiology
68,571-6.
Matyssek R, Mamyama S, Boyer JS. 1988. Rapid wall
relaxation in elongating tissues. Plant Physiology 86, 1163-7.
McCully ME. 1995. How do real roots work? Some new views
of root structure. Plant Physiology 109, 1-6.
McCully ME, Canny MJ. 1988 Pathways and processes of
water and nutrient movement in roots. Plant and Soil
111, 159-70.
Melchior W, Steudle E. 1993. Water transport in onion (Allium
cepa L.) roots. Changes of axial and radial hydraulic
conductivities during root development. Plant Physiology
101, 1305-15.
Mesbcheryakov A, Steudle E, Komor E. 1992. Gradients of
turgor, osmotic pressure, and water potential in the cortex of
the hypocotyl of growing Ricinus seedlings. Effects of the
supply of water from the xylem and of solutes from
the phloem. Plant Physiology 98, 840-52.
Michelena VA, Boyer JS. 1982. Complete turgor maintenance
at low water potentials in the elongating region of maize
leaves. Plant Physiology 69, 1145-9.
Mizuno A, Nakahori K, Katou K. 1993. Acid-induced changes
in in vivo wall yielding properties of hypocotyl sections of
Vigna unguiculata. Physiologia Plantarum 89, 693-8.
Munns R. 1988. Why measure osmotic adjustment? Australian
Journal of Plant Physiology 15, 717-26.
Nonami H, Boyer JS. 1993. Direct demonstration of a growthinduced water potential gradient. Plant Physiology 102, 13-19.
Okamoto H, Liu Q, Nakahori K, Katou K. 1989. A pressurejump method as a new tool in growth physiology for
monitoring physiological wall extensibility and effective
turgor. Plant and Cell Physiology 30, 979-85.
Okamoto H, Okamoto A. 1994. The pH-dependent yield
threshold of the cell wall in a glycerinated hollow cylinder (in
vitro system) of cowpea hypocotyl. Plant, Cell and
Environment 17, 979-83.
Oparka KJ, Prior DAM, Wright KM. 1995. Symplastic
communication between primary and developing lateral roots
of Arabidopsis thaliana. Journal of Experimental Botany
46, 187-97.
Oparka KJ, Wright KM. 1988. Influence of cell turgor on
sucrose partitioning in potato tuber storage tissues. Planta
175, 520-6.
Ortega JKE. 1985. Augmented growth equation for cell wall
expansion. Plant Physiology 79, 318-20.
Passioura JB. 1988. Water transport in and to roots. Annual
Review of Plant Physiology 39, 245-65.
Passioura JB, Fry SC. 1992. Turgor and cell expansion: beyond
the Lockhart equation. Australian Journal of Plant Physiology
19, 565-76.
Root elongation and drought
Patrick JW. 1994. Turgor-dependent unloading of photosynthates from coats of developing seed of Phaseolus vulgaris
and Viciafaba. Turgor homeostasis and set points. Phvswlogia
Plantarum 90, 367-77.
Peterson CA, Steudle E. 1993. Lateral hydraulic conductivity of
early metaxylem vessels in Zea mays L. roots. Planta
189, 288-97.
Pockman WT, Sperry JS, O'Leary JW. 1995. Sustained and
significant negative water pressure in xylem. Nature 378,
715-16.
Pritchard J, Barlow PW, Adam JS, Tomos AD. 1990. Biophysics
of the inhibition of the growth of maize roots by lowered
temperature. Plant Physiology 93, 222-30.
Pritchard J, Hetherington PR, Fry SC, Tomos AD. 1993.
Xyloglucan endotransglycosylase activity, microfibril orientation and the profiles of cell wall properties along growing
regions of maize roots. Journal of Experimental Botany
44, 1281-9.
Pritchard J, Wyn Jones RG, Tomos AD. 1991. Turgor, growth
and rheological gradients of wheat roots following osmotic
stress. Journal of Experimental Botany 42, 1043-9.
Rail W. 1987. Neuron, cable properties. In: Adelman G, ed.
Encyclopedia of Neuroscience. Boston: BirkhSuser Inc.,
816-20.
Sanderson J. 1983. Water uptake by different regions of the
barley root. Pathways of radial flow in relation to development
of the endodermis. Journal of Experimental Botany 34, 240-53.
Schulz A. 1995. Plasmodesmal widening accompanies the short
term increase in symplasmic phloem unloading in pea root
tips under osmotic stress. Protoplasma 188, 22-37.
Schulze E-D. 1986. Carbon dioxide and water vapor exchange
in response to drought in the atmosphere and in the soil.
Annual Review of Plant Physiology 37, 247-74.
Serpe MD, Matthews MA. 1992. Rapid changes in cell wall
yielding of elongating Begonia argenteo-guttata L. leaves in
response to changes in plant water status. Plant Phvsiologv
100, 1852-7.
Shackel KA, Matthews MA, Morrison JC. 1987. Dynamic
relation between expansion and cellular turgor in growing
grape (Vitis vinifera L.) leaves. Plant Physiology 84, 1166-71.
Sharp RE, Hsiao TC, Silk WK. 1990. Growth of the maize
primary root at low water potentials. II. The role of growth
and deposition of hexose and potassium in osmotic adjustment. Plant Physiology 93, 1337^6.
Sharp RE, Silk WK, Hsiao TC. 1988. Growth of the maize
primary root at low water potentials. I. Spatial distribution
of expansive growth. Plant Physiology 87, 50-7.
Silk WK. 1984. Quantitative descriptions of development.
Annual Review of Plant Physiology 35, 479-518.
Silk WK, Wagner KK. 1980. Growth-sustaining water potential
distributions in the primary corn root. A noncompartmented
continuum model. Plant Physiology 66, 859-63.
999
Slovik S, Daeter W, Hartung W. 1995. Compartmental redistribution and long-distance transport of abscisic acid (ABA)
in plants as influenced by environmental changes in the
rhizosphere—a
biomathematical
model.
Journal of
Experimental Botany 46, 881-94.
Spollen WG, Sharp RE. 1991. Spatial distribution of turgor and
root growth at low water potentials. Plant Physiology
96, 438-43.
Steudle E. 1989. Water flow in plants and its coupling to other
processes: an overview. In: Fleischer S, Fleischer R, eds.
Methods in enzymology, No. 174. Biomembranes, Part U:
Cellular and subcellular transport: Eukaryotic (non-epithelial)
cells. London: Academic Press, 183-225.
Steudle E. 1992. The biophysics of plant water: compart mentation, coupling with metabolic processes, and flow of
water in plant roots. In: Somero GN, Osmond CB, Bolis CL,
eds. Water and life: comparative analysis of water relationships
at the organismic, cellular, and molecular levels. Heidelberg,
Germany: Springer-Verlag, 173-204.
Steudle E. 1994. Water transport across roots. Plant and Soil
167, 79-90.
Steudle E. 1995. Trees under tension. Nature 378, 663-4.
Steudle E, Frensch J. 1989. Osmotic responses of maize roots:
water and solute relations. Planta 177, 281-95.
Steudle E, Frensch J. 1996. Water transport in plants: role of
the apoplast. Plant and Soil, 187, 67-79.
Tardieu F, Davies WJ. 1993. Integration of hydraulic and
chemical signalling in the control of stomatal conductance
and water status of droughted plants. Plant, Cell and
Environment 16, 341-9.
Taylor RE. 1963. Cable theory. In: Nastuk WL, ed. Physical
techniques in biological research. IV. Electrophysiological
methods, Part B. New York: Academic Press, 219-62.
Tomos AD, Malone M, Pritchard J. 1989. The biophysics of
differential growth. Environmental and Experimental Botany
29, 7-23.
Tomos AD, Pritchard J. 1994. Biophysical and biochemical
control of cell expamsion in roots and leaves. Journal of
Experimental Botany 45, 1721-31.
Waisel Y, Eshel A, Kalkafi U. 1991. Plant roots: the hidden half.
New York: Marcel Decker.
Zhu GL, Steudle E. 1991. Water transport across maize roots.
Simultaneous measurement of flows at the cell and root level
by double pressure probe technique. Plant Physiology 95,
305-15.
Zimmcrmann U, Haase A, Langbein D, Meinzer F. 1993.
Mechanisms of long-distance water transport in plants: a
re-examination of some paradigms in the light of new
evidence. Philosophical Transaction of the Royal Society of
London, Series B 341, 19-31.