Journal of Experimental Botany, Vol. 51, No. 343, pp. 239–248, February 2000
The influence of supra-optimal root-zone temperatures
on growth and stomatal conductance in Capsicum
annuum L.
I.C. Dodd1,3, J. He2, C.G.N. Turnbull1, S.K. Lee2 and C. Critchley1
1 Department of Botany, The University of Queensland, St Lucia 4072, Australia
2 School of Science, Nanyang Technological University, 469 Bukit Timah Road, Singapore 259756
Received 4 May 1999; Accepted 24 September 1999
Abstract
Introduction
Pepper (Capsicum annuum L.) plants were grown aeroponically in a Singapore greenhouse under natural
diurnally fluctuating ambient shoot temperatures, but
at two different root-zone temperatures (RZTs): a constant 20±2 °C RZT and a diurnally fluctuating ambient
(A) (25–40 °C) RZT. Plants grown at 20-RZT had more
leaves, greater leaf area and dry weight than A-RZT
plants. Reciprocal transfer experiments were conducted between RZTs to investigate the effect on plant
growth, stomatal conductance (g ) and water relations.
s
Transfer of plants from A-RZT to 20-RZT increased
plant dry weight, leaf area, number of leaves, shoot
water potential (Y
), and g ; while transfer of plants
shoot
s
from 20-RZT to A-RZT decreased these parameters.
Root hydraulic conductivity was measured in the latter
transfer and decreased by 80% after 23 d at A-RZT.
Transfer of plants from 20-RZT to A-RZT had no effect
on xylem ABA concentration or xylem nitrate concentration, but reduced xylem sap pH by 0.2 units. At both
RZTs, g measured in the youngest fully expanded
s
leaves increased with plant development. In plants with
the same number of leaves, A-RZT plants had a higher
g than 20-RZT plants, but only under high atmospheric
s
vapour pressure deficit. The roles of chemical signals
and hydraulic factors in controlling g of aeroponically
s
grown Capsicum plants at different RZTs are discussed.
Ambient tropical conditions of high temperatures and
light intensities severely reduce growth of temperate vegetable crops if root-zone temperature (RZT ) is not controlled (Lee and Cheong, 1996; He and Lee, 1998a, b).
Potential mechanisms of growth inhibition at ambient
RZT include poor nutrient uptake and/or perturbed water
relations.
Although bulk tissue nutrient concentration can indicate RZT-induced nutrient deficiency (Sheppard et al.,
1986), methods which maximize the contribution of the
xylem to the sample, such as petiolar (Olsen and Lyons,
1994) or stump (Barthes et al., 1996) sap extracts, may
be better predictors of nutrient status. Xylem sap nitrate
concentration ([ X-NO−]) may be an important indicator
3
of perturbed nutrient relations for several reasons. Nitrate
is a major xylem sap solute (Bassirirad et al., 1991) and
leaf growth responds sensitively and rapidly to any perturbation in nitrate supply (Palmer et al., 1996). Changes
in [ X-NO−] will reflect changes in root uptake as there
3
is minimal recirculation in the phloem (Jeschke and Pate,
1992). Furthermore, nitrate uptake may be particularly
sensitive to elevated RZT in some species ( Udagawa
et al., 1991).
Supra-optimal RZT can induce shoot water deficit by
altering the balance between water uptake by the root
system and water loss from the shoots. Transfer of roots
to a supra-optimal temperature transiently (up to 48 h;
Newman and Davies, 1988) increases root hydraulic
conductivity (Lp). However, in the long-term (e.g. 28 d;
Key words: Capsicum, root temperature, growth, hightemperature stress, stomata, chemical signalling.
3 To whom correspondence should be addressed. Fax: +61 7 3365 1699. E-mail: [email protected]
Abbreviations: A-RZT, ambient root-zone temperature; A[20-RZT, plants grown initially at A-RZT then transferred to 20-RZT; g , stomatal conductance;
s
Lp, root hydraulic conductivity; RZT, root-zone temperature; Y , leaf water potential; Y
, shoot water potential; 20-RZT, root-zone temperature of
leaf
shoot
20±2 °C; 20[A-RZT, plants grown initially at 20-RZT then transferred to A-RZT; VPD, atmospheric vapour pressure deficit; [X-ABA], xylem sap ABA
concentration; [X-NO− xylem sap nitrate concentration.
3
© Oxford University Press 2000
240 Dodd et al.
Graves et al., 1991) Lp is decreased. When water loss
exceeds water uptake, the resultant declines in leaf turgor
can directly restrict leaf growth (Bunce, 1977) and close
stomata ( Turner, 1974), resulting in loss of photosynthetic productivity.
However, variability in stomatal responses and shoot
water relations under supra-optimal RZT suggests that
altered shoot water relations may not entirely account
for the observed stomatal behaviour. Certainly, there are
cases where a concurrent reduction of both shoot water
potential ( Y
) and stomatal conductance ( g ) (Graves
shoot
s
and Aiello, 1997) suggests that stomata are responding
directly to reduced Y
or reduced turgor. Reduced
shoot
Y without stomatal closure (Graves et al., 1991; Menzel
leaf
et al., 1994) may result from osmotic adjustment maintaining shoot turgor. Stomatal closure without reduced
Y
(Graves et al., 1991; Behboudian et al., 1994) has
leaf
been interpreted as evidence for root-derived chemical
signals moving via the xylem to the shoots to reduce g .
s
A principal candidate for such a signal is the plant
hormone abscisic acid (ABA) (reviewed in Dodd et al.,
1996). Recently, it has been shown that alkalization of
the xylem sap, without increased xylem sap ABA concentration ([ X-ABA]), can cause stomatal closure
( Wilkinson et al., 1998). However, as far as is known,
there are no reports measuring either [ X-ABA] or xylem
pH in plants grown at supra-optimal RZTs.
Previous work has shown significant effects of manipulating RZT on lettuce growth in the tropics (Lee and
Cheong, 1996; He and Lee, 1998a, b). However, investigation of chemical signalling in lettuce is impractical, due
to the difficulty of obtaining a xylem sap sample free of
contamination from the latex which exudes from cut
lettuce stumps. Xylem sap extraction from pepper
(Capsicum annuum L.) shoots using a pressure bomb
presented no such problems. Pepper growth at different
RZTs has not been evaluated previously under tropical
aerial conditions. However, at shoot temperatures of
23/19 °C (day/night), pepper showed increased growth
with increasing root temperatures up to 30 °C, but then
a negative response of leaf area production and dry
weight at 36 °C RZT (Gossellin and Trudel, 1986). It was
hypothesized, therefore, that pepper would show reduced
growth and g at ambient RZTs in Singapore. It was also
s
evaluated whether shoot water relations and/or xylem sap
composition were altered at different RZTs.
Materials and methods
Plant material and growth conditions
Experiments 1 and 2: Experiments were conducted using
aeroponics units (described by Lee and Cheong, 1996) maintained in a greenhouse at Nanyang Technological University,
Singapore ( latitude 2° N ) during April–May 1997. Pepper
(Capsicum annuum L. cv. Jin Jiao No. 3) seeds were germinated
in seedling trays. Two batches of seedlings were germinated at
a 2 week interval, and transferred to water-soaked polyurethane
cubes 20 d after sowing. Three days later, the foam cubes were
transferred to aeroponics units. Nutrient solution (full-strength
Netherlands Standard Composition; Douglas, 1982) was misted
over the roots for one 20 s burst every 90 s. Solution conductivity
was maintained at 2.2 mS.
When a data logger (Data Hog, Skye, Wales) was available,
temperature (thermistor) and relative humidity ( Vaisala-type
sensor) were recorded inside the greenhouse every 10 min.
During periods of logger unavailability, greenhouse data were
estimated from empirical relationships constructed between
greenhouse conditions and those at a nearby Singapore
Department of Environment meteorological station. The aerial
parts of all plants were exposed to ambient greenhouse
temperatures and vapour pressure deficits ( VPDs). Average
maximum greenhouse temperatures and VPDs during the
experimental period were 37.0±0.7 °C and 3.9±0.2 kPa,
respectively. The average minimum greenhouse temperature
was 26.4±0.2 °C. Half the plants were maintained at a constant
RZT of 20±2 °C while the remainder were exposed to a
diurnally fluctuating ambient RZT (A-RZT ). On sunny days,
temperature of the nutrient solution at A-RZT remained at
40±2 °C between 11.00 h and 18.00 h. Incident quantum flux
at the canopy surface reached maxima of c. 1700 mmol m−2 s−1.
Experiment 3: An additional experiment was conducted in
September 1998 with Capsicum annuum cv. Indra F1 hybrid.
Plant culture was as previously described. Several batches of
A-RZT and 20-RZT plants were transplanted into the aeroponics units over several weeks to produce plants grown at different
RZTs, but of a similar developmental stage at the time of
measurement. Average maximum greenhouse temperatures and
VPDs during the experimental period were 38.7±0.8 °C and
4.2±0.3 kPa, respectively. The average minimum greenhouse
temperature was 24.9±0.2 °C.
Physiological measurements
Root-zone temperature transfers were conducted 55 d
( Experiment 1) and 41 d ( Experiment 2) after sowing, using
two separate batches of plants. Half the plants of each batch
were maintained at their original RZT (either 20- or A-RZT ),
while the remainder were transferred to the other RZT at
07.30 h under diffuse light (<100 mmol m−2 s−1). There were
thus four RZT treatments: 20-RZT, A-RZT, 20[A-RZT, and
A[20-RZT.
Every 3–5 d over a 25 d measurement period (5 d before and
20 d after the temperature transfer), Y
and g were detershoot
s
mined between 09.00 h and 15.00 h. Stomatal conductance of
the three youngest fully expanded leaves ( g ) was measured
s
using a porometer (AP4, Delta-T Devices, Burwell, UK ). Small
leaf size of A-RZT plants in Experiment 2 prevented porometric
determination of g until day 4. Xylem sap was sampled from
s
the same plants used for g and water potential measurements.
s
The water potential (Y
) of the entire shoot (decapitated
shoot
above the cotyledons) was determined using a pressure bomb
(Plant Moisture Systems, Santa Barbara, CA, USA) whose
chamber was lined with moistened filter paper. Following
measurement of Y
in plants originally grown at 20-RZT,
shoot
xylem sap was collected at 0.5 MPa above the balancing
pressure for 5 min for subsequent analysis of xylem sap
constituents. Small size of plants originally grown at A-RZT
prevented the collection of sufficient xylem sap for analysis.
After these measurements, the number of leaves (>10 mm
long), total leaf area and total dry weight of each plant were
determined.
Growth, stomatal conductance and root temperature 241
On one occasion in Experiment 2, root hydraulic conductivity
(Lp) was estimated by adapting a method used previously (Ford
and Harrison-Murray, 1997). Plants were decapitated just
below the cotyledons, and the root system plus stump was
placed in a beaker of nutrient solution in the pressure bomb.
After the plant was sealed into the chamber, the pressure was
raised in 0.1 MPa increments (to a maximum of 0.6 MPa) and
held for 3 min at each pressure to allow collection of the
exudate by a capillary tube. The slope of the relationship
between exudate flow rate (in mm3 s−1) and applied pressure
over the linear part of the curve (0.2–0.6 MPa) gave the root
hydraulic conductivity (Lp). In all cases, correlation coefficients
(r2) were >0.95. The temperature of the nutrient solution was
determined before and after pressure application, and did not
deviate by more than 3 °C during this time. Removal of the
root system in preliminary tests confirmed that the major
resistance to water flow was in the roots, and not the stem.
Experiment 3 measured g of leaves of 20-RZT and A-RZT
s
plants at multiple points of insertion on the main stem on
sunny and cloudy days. Y
was also measured over the
shoot
course of a sunny day.
Measurement of xylem sap constituents
Xylem ABA concentration was measured by gas chromatography-mass spectrometry (GC-MS). Twenty nanograms of
[2H ] ABA were added to 40 mm3 of sap. Sap was evaporated
6
to dryness under vacuum, redissolved in 20 mm3 of methanol,
then diazomethane added to methylate ABA. Samples were
then dried, redissolved in 20 mm3 of ethyl acetate, and 2 mm3
of sample injected onto a fused silica DB 5 MS capillary column
(25 m×0.2 mm×0.33 mm) with a 5% phenyl, 95% methyl
silicone stationary phase coating (J & W Scientific, Folsom CA,
USA). The ion ratios 190:194 and 162:166 were used to
calculate the ABA concentration of duplicate samples.
Xylem nitrate concentration was measured in 20 mm3 of sap
by the nitrite colorimetric reaction with cadmium as the
reducing agent. After 3 min of reaction, 1000 mm3 of the
reaction mixture was added to 500 mm3 of sulphanilic acid and
500 mm3 of a-naphthyl ethylene diamine dihydrochloride. This
mixture was incubated for 30 min before absorbance at 540 nm
was determined using a spectrophotometer (Model U-1100,
Hitachi).
Xylem pH was measured with a microelectrode (Model
98-02 BN, Orion Instruments, Boston, MA, USA) interfaced
with a pH meter (Model 900I3, TPS Ionode, Brisbane,
Australia).
Statistics
Student’s t-test was used to discriminate differences between
plants which were transferred to a new RZT and those that
remained at their original RZT. The significance of regressions
was tested in JMP In (SAS Institute Inc., Cary, NC, USA).
Results
On the day prior to the RZT transfers, there were
significant (P<0.05) differences in the number of leaves
per plant, total leaf area, and total dry weight of plants
grown at different RZTs ( Table 1). Total leaf area and
total dry weight of A-RZT plants was c. 10% of 20-RZT
plants. The effect on leaf number was much less, presumably due to the initiation of leaves in the period between
germination and transfer of the plants to the growth units.
Transfer of plants from A-RZT to 20-RZT generally
increased total dry weight ( Fig. 1a, b), total leaf area
( Fig. 1c, d) and number of leaves ( Fig. 1e, f ). This
response was quite slow to occur, presumably because
the plants had yet to reach their exponential growth
phase. In contrast to this sluggish response, transfer of
plants from 20-RZT to A-RZT decreased total dry weight
( Fig. 2a, b), total leaf area (Fig. 2c, d) and number of
leaves (Fig. 2e, f ) within 8–13 d. In Experiment 1, this
resulted in a 33% reduction in total dry weight after 18 d
( Fig. 2a), and a 29% reduction in both total leaf area
( Fig. 2c) and number of leaves ( Fig. 2e) after 13 d.
Reduced leaf area expansion and initiation between 13 d
and 18 d in 20-RZT plants was due to fruiting of these
plants. In Experiment 2, the RZT transfer resulted in a
38% reduction in total dry weight after 13 d ( Fig. 2b), a
41% reduction in total leaf area after 13 d (Fig. 2d ), and
a 23% reduction in number of leaves after 8 d ( Fig. 2f ).
In both experiments, Y
of A-RZT and A[20-RZT
shoot
plants increased as the plants grew (Fig. 1i, j). Transferred
plants showed significantly (P<0.01) higher Y
on
shoot
day 8 in Experiment 1 (Fig. 1i), and day 18 in Experiment
2 (Fig. 1j). On both occasions, stomatal conductance of
transferred plants was also significantly (P<0.05) higher,
and also on day 17 in Experiment 1 (Fig. 1k), and day
13 in Experiment 2 (Fig. 1l ).
Y
of 20-RZT plants and 20[A-RZT plants did
shoot
not statistically differ (P>0.05) except on day 17 in
Experiment 2 ( Fig. 2j), when Y
of 20[A-RZT plants
shoot
was 0.18 MPa lower than 20-RZT plants. Regression
analysis of data collected after the transfer (i.e. after day
0) showed no significant (P>0.05) effect of time on Y
shoot
( Fig. 2i, j). There were four occasions when stomatal
conductance of 20[A-RZT plants was lower than that
of 20-RZT plants (Fig. 2k, l ).
In 20-RZT and 20[A-RZT plants, the shoot was large
enough to permit collection of sufficient xylem sap for
analysis throughout the transfer experiments. For all data
collected after the transfer (i.e. after day 0) in both
experiments, regression analysis showed no significant
(P>0.05) effect of time on xylem sap ABA concentration,
xylem nitrate concentration or xylem sap pH (data not
shown). Accordingly, data from all harvests were combined to give mean values at each RZT ( Table 2). Y
,
shoot
[ X-ABA] and [ X-NO−] in both experiments, and xylem
3
pH in Experiment 1, did not significantly (P>0.05) differ
between 20-RZT and 20[A-RZT plants. In Experiment
2, xylem pH of 20[A-RZT plants was significantly
(P<0.05) less (0.2 units) than plants that remained at
20-RZT, a difference of similar magnitude to that seen in
Experiment 1.
Root hydraulic conductivity (Lp) of 20-RZT and
20[A-RZT plants was measured 23 d after the RZT
transfer in plants remaining from Experiment 2.
Measurements occurred at two nutrient solution temperatures, 20 °C and 30 °C, to allow for temperature effects
242 Dodd et al.
Table 1. Growth indices of pepper plants 55 d (Experiment 1) and 41 d (Experiment 2) after germination
From 23 d after germination, plants were grown at either 20-RZT or ambient RZT. Values are mean±SE of four replicates.
Variable
Experiment 1
Number of leaves
Leaf area (cm2)
Total dry wt. (g)
Experiment 2
Number of leaves
Leaf area (cm2)
Total dry wt. (g)
20-RZT
Ambient RZT
Ambient as % of 20-RZT
13.8±1.0
86±23.5
0.656±0.191
8.3±0.3
9.0±2.5
0.068±0.021
60.4
10.5
10.4
11.0±0.7
26±4.4
0.233±0.058
5.5±0.5
3.5±0.5
0.026±0.004
50.0
13.5
11.0
Fig. 1. Changes in total plant dry weight (a, b), total plant leaf area (c, d), number of leaves per plant (e, f ), shoot water potential ( Y
) (i, j),
shoot
and stomatal conductance (g ) (k, l ) of plants grown and maintained at A-RZT (,) and those grown at A-RZT but transferred to 20-RZT (V ) at
s
day 0. Average atmospheric VPD in the greenhouse between 09.00 h and 15.00 h on the measurement occasions is given (g, h). Stomatal
conductance measurements are mean±SE of 12 leaves on four plants, while dry weight, leaf area, number of leaves and Y
are mean±SE of
shoot
four plants. Treatment differences as determined by Student’s unpaired t-test as indicated: P<0.10 *, P<0.01 **, P<0.001 ***.
Growth, stomatal conductance and root temperature 243
Fig. 2. Changes in total plant dry weight (a, b), total plant leaf area (c, d), number of leaves per plant (e, f ), shoot water potential ( Y
) (i, j),
shoot
and stomatal conductance (g ) (k, l ) of plants grown and maintained at 20-RZT (#) and those grown at 20-RZT but transferred to A-RZT ($)
s
at day 0. Average atmospheric VPD in the greenhouse between 09.00 h and 15.00 h on the measurement occasions is given (g, h). Stomatal
conductance measurements are mean±SE of 12 leaves on four plants, while dry weight, leaf area, number of leaves and Y
are mean±SE of
shoot
four plants. Treatment differences as determined by Student’s unpaired t-test as indicated: P<0.10 *, P<0.01 **, P<0.001 ***.
Table 2. Xylem sap composition and average Y
of pepper plants between 0 and 18 d after the root-zone temperature transfers
shoot
Xylem sap values are mean±SE of 10–16 replicates (xylem sap collected from four plants on four occasions, but insufficient sap for analysis of all
variables in some samples). P values from Student’s unpaired t-test are indicated.
Variable
Experiment 1
Experiment 2
Treatment
Y
(MPa)
shoot
Xylem ABA (nM )
Xylem nitrate (mM )
Xylem pH
P value
20-RZT
20[A-RZT
0.94±0.05
45±4
0.79±0.18
6.54±0.14
1.03±0.06
46±3
0.81±0.10
6.34±0.14
0.22
0.94
0.95
0.32
Treatment
P value
20-RZT
20[A-RZT
0.77±0.05
65±6
0.51±0.11
6.55±0.05
0.88±0.05
52±4
0.40±0.06
6.33±0.05
0.14
0.09
0.53
0.013
244 Dodd et al.
However, there was no difference in g between A-RZT
s
and 20-RZT plants for any leaf. On the sunny day
(average VPD during measurement=3.0 kPa), g of plants
s
at both RZTs was lower than on the low VPD day. On
this occasion, 20-RZT plants were wilted and had a g
s
only 30–45% of A-RZT plants. On another sunny day
(2 d prior to the g measurements), Y
of 20-RZT
s
shoot
plants (−1.08±0.04 MPa, n=8) was much less than
Y
of A-RZT plants (−0.78±0.02 MPa, n=7).
shoot
Discussion
Fig. 3. Relationship between pressure-induced exudate flow and applied
pressure for four plants 23 d after a RZT transfer. Measurements were
made between 09.00 h and 11.00 h. Linear regressions fitted in SigmaPlot
for Windows 2.01; the slope of the line being the root hydraulic
conductivity (Lp).
on the viscosity of water. Irrespective of nutrient solution
temperature at the time of Lp determination, Lp of
20[A-RZT plants was only 20% of 20-RZT plants ( Fig.
3). This difference remained when data were normalized
for different root dry weights between plants.
To examine relationships between g , Y
and number
s shoot
of leaves, only data for sunny days (average VPD between
09.00 h and 15.00 h >2.5 kPa) were considered (Figs 1g,
h; 2g, h). Data were grouped according to the RZT at
which the plants were originally grown. The significance
of regressions for various data sets is given in Table 3.
For both groups of plants, g increased with an increasing
s
number of leaves per plant. The slope of this relationship
differed according to the RZT at which the plants were
originally grown (Fig. 4a). Between 10 and 15 leaves per
plant, plants originally grown at A-RZT maintained a
much higher g . Y
also increased with the number of
s shoot
leaves, but only in plants that were originally grown at
A-RZT (Fig. 4b). However, analysis of the data sets
from both groups of plants showed no relationship
between Y
and the number of leaves when data for
shoot
plants with <8 leaves were excluded. Stomatal conductance was correlated with Y
, but only in plants that
shoot
were originally grown at A-RZT (Fig. 4c).
Further experiments sought to confirm the effect of
RZT on g at two different VPDs, in plants of the same
s
developmental stage. On the cloudy day (average VPD
during measurement=1.5 kPa), g of plants at both RZTs
s
increased with leaf insertion level on the main stem.
Capsicum annuum is chilling sensitive and night temperatures less than 20 °C limit leaf initiation and growth
(Mercado et al., 1997). Although the 20-RZT treatment
may therefore be suboptimal, in these experiments it still
greatly increased growth compared to A-RZT plants
( Table 1). Thus Capsicum responds similarly to lettuce
(Lee and Cheong, 1996; He and Lee, 1998a, b) although
further experiments conducted simultaneously under
identical aerial conditions have shown that dry matter
accumulation at A-RZT is more impaired in lettuce than
pepper (J He and SK Lee, unpublished observations).
Although the growth of pepper plants at 20-RZT was
much greater than at A-RZT, an optimal RZT for pepper
under tropical aerial conditions is yet to be determined.
The RZT transfer experiments generally showed coordinate changes in total dry weight, leaf area and number
of leaves ( Figs 1, 2). Leaf area development is dependent
on three component processes: individual leaf area expansion, the initiation of new leaves, and initiation of new
axillary shoots. In Capsicum, the accumulation of leaf
number after the appearance of a set number of mainstem leaves (12 in our experiments) is dependent on
monopodial branching. This branching process was particularly responsive to RZT (Fig. 2e, f; TW Choong, J
He and SK Lee, unpublished observations). Individual
leaf area at specific nodes was not determined, which
would have discriminated whether RZT affected leaf
expansion per se, or simply the rate of plant development.
Future studies should aim to determine the sensitivity of
the component processes to high RZT stress.
Delayed leaf area development could be a response to
either water stress or suboptimal nitrogen nutrition.
However, no differences were found in xylem nitrate
concentrations following transfer of plants from 20-RZT
to A-RZT ( Table 2). This contrasts with previous data
which found that exposure of sorghum roots to 35 °C for
only 4 h reduced [ X-NO−] by 30% (Bassirirad et al.,
3
1991). Although [ X-NO−] was unaffected by the RZT
3
transfer in these experiments, it is possible that NO− flux
3
to the shoots was decreased due to stomatal closure
reducing transpirational flow. Since large reductions in
leaf area preceded stomatal closure (cf. Fig. 2c, d and
Fig. 2k, l ), it seems unlikely that decreased nitrate flux
was the primary cause of reduced leaf area development.
Growth, stomatal conductance and root temperature 245
Although reduced total leaf area may partially explain
the reduction in total dry weight of 20[A-RZT plants,
so may reductions in photosynthetic rate of specific leaves.
Photosynthesis at A-RZT may be reduced by both photoinhibition ( He and Lee, 1998a, b) and stomatal closure.
In this paper, the regulation of stomatal conductance was
investigated.
Analysis of changes in g between specific RZT treats
ments in the transfer experiments ( Figs 1, 2) was complicated by varying environmental conditions (Fig. 1g, h)
and plant ontogeny ( Fig. 4a). The effect of plant development on g was especially striking considering that the
s
youngest fully expanded leaves were used in all measurements. Other species do not show increases in maximum g
s
with increasing node number (e.g. sunflower; Zhang and
Davies, 1989), but in Capsicum, interpretation of differences in g between treatments must consider the effects
s
of measuring leaves at different node numbers. In the ARZT to 20-RZT transfer, there were minimal effects of
the new RZT on leaf number ( Fig. 1e, f ), thus measurement of plants at different developmental stages is unlikely
to account fully for the increased g of A[20-RZT plants.
s
However, the decreased g of 20[A-RZT plants (relative
s
to 20-RZT plants) was first detected when these plants
had fewer leaves (cf. Fig. 2e, f and Fig. 2k, l ), and thus
measurement of plants at different developmental stages
may have contributed to the measured differences in g .
s
However, when leaf number in the 20-RZT and 20[ARZT plants was more similar at the end of the experiment,
an alternative explanation was needed for stomatal
closure.
Consequently, it was necessary to use plants of the
same developmental stage to study the effects of atmospheric vapour pressure deficit ( VPD) and RZT on g
s
(Fig. 5). Stomatal response to either variable was dependent on the other. At low VPDs, g of A-RZT and 20s
RZT plants were similar. Under high VPDs, 20-RZT
plants wilted, resulting in stomatal closure. Wilting at
chilling temperatures (of whole plants with both roots
and shoots at the same temperature) has previously been
shown to depend on atmospheric VPD (McWilliam et al.,
1982). Low temperature induced reductions in root
hydraulic conductivity (Lp) (Markhart et al., 1979) result
in water uptake being unable to keep pace with transpir-
ation. Although decreased Lp of 20-RZT plants would
be expected simply due to the lower viscosity of water at
this temperature, differences in root morphology between
roots grown at different temperatures may also have
contributed.
Stomatal behaviour in the RZT transfer experiments
may also have been affected by changes in Lp. Reduced
Lp may lower Y
, which in turn can cause stomatal
shoot
closure. Certainly, in both RZT transfer experiments
there were measurement occasions when altered Y
shoot
was temporally associated with altered stomatal behaviour. Greater stomatal conductance of A[20-RZT plants
(relative to A-RZT plants) was generally associated with
increased Y
(cf. Fig. 1i, j and Fig. 1k, l ), implying a
shoot
hydraulic limitation of g in A-RZT plants. This limitation
s
is unlikely to result from high temperature per se (since
Lp increases with increasing temperature), but via an
effect of high RZT on root development. During the early
stages of growth, Lp rapidly increases with development
( Fiscus and Markhart, 1979), which would explain the
observation that Y
increased with plant development
shoot
in plants originally grown at A-RZT ( Fig. 1i, j). Although
Lp was not assessed in these experiments, the increased
root growth of A[20-RZT plants (data not shown) may
have increased Lp (even though root temperature was
lower), allowing improved shoot water status.
When 20-RZT plants were transferred to A-RZT, a
large reduction in Lp was measured after 23 d (Fig. 3).
Roots of these plants appeared swollen and brown,
indicating that they were unable to acclimate to their new
root temperature. Despite this large reduction in Lp, the
decrease in g of 20[A-RZT plants following the RZT
s
transfer was associated with decreased Y
on only one
shoot
occasion at the end of Experiment 2 ( Fig. 2j). Since
Y
did not change with time after the transfer in both
shoot
20-RZT and 20[A-RZT plants, it was possible to average
Y
data across all measurement occasions. However,
shoot
this still failed to yield significant (P<0.10) differences in
Y
( Table 2). Thus it was necessary to examine the
shoot
possibility that altered xylem sap composition was
responsible for stomatal closure in 20[A-RZT plants.
Although increased xylem ABA concentration is often
well correlated with stomatal closure (reviewed in Dodd
et al., 1996), no change in [ X-ABA] was detected after
Table 3. Significance of regressions presented in Fig. 4
Regressions were calculated using mean values at each measurement occasion. P values of each regression are indicated.
Regressions
g on Number of Leaves (Fig. 4a)
s
Y
on Number of Leaves (Fig. 4b)
shoot
g on Y
( Fig. 4c)
s
shoot
Regression type
Linear
Linear
Linear
Data set
All plants
P value
20-, 20[ARZT plants
P value
A-, A[20RZT plants
P value
0.0015
0.256
0.141
0.0004
0.0003
<0.0001
0.0012
0.007
<0.0001
246 Dodd et al.
Fig. 5. Effect of main stem node number on the stomatal conductance
of the leaf at that node for 20-RZT (#) and A-RZT (V ) plants on a
sunny day, and for 20-RZT ($) and A-RZT (,) plants on a cloudy
day. In both cases the leaf of highest node number measured was the
youngest fully expanded leaf. Data are mean±SE of at least five plants.
Fig. 4. Relationships between stomatal conductance and number of
leaves (a), Y
and number of leaves (b), and stomatal conductance
shoot
and Y
(c) for plants grown and maintained at 20-RZT (#) or at
shoot
A-RZT (,), or grown at 20-RZT but transferred to A-RZT ($) at
day 0, or grown at A-RZT but transferred to 20-RZT (V ) at day 0.
Data are from Figs 1 and 2. Error bars in (c) omitted for clarity. Lines
are significant (P<0.05) linear regressions fitted in SigmaPlot for
Windows 2.01. In (b) and (c), lines fitted only to the group of A-RZT
and A[20-RZT plants.
transfer of 20-RZT plants to A-RZT ( Table 2). Although
increases in ABA flux of an order of magnitude may
cause stomatal closure independent of changes in ABA
concentration ( Trejo et al., 1995), the reduced g of
s
20[A-RZT plants implied that ABA flux was also
decreased. Recent studies have considered a role for
stomatal sensitivity to xylem-borne ABA in stomatal
closure (Dodd et al., 1996), and a possible effect of RZT
on this sensitivity should be assessed before entirely
dismissing a role for ABA in stomatal closure in 20[ARZT plants.
Alkalization of the xylem sap is a common response to
many root stresses and can cause stomatal closure at [ XABA]s typical of those found in well-watered plants
( Wilkinson et al., 1998). High RZT stress did alter xylem
sap pH ( Table 2), but in the opposite direction from that
required to close the stomata. There are cases where
droughted plants show reduced xylem sap pH (Schurr
and Schulze, 1996) thus acidification of xylem sap in
response to a root stress is not without precedent. Clearly
the regulation of xylem sap pH as a stress signal requires
further investigation.
Reduced xylem sap cytokinin flux has also been implicated in stomatal closure (Shashidhar et al., 1996) and it
is known that supra-optimal RZT can reduce root cytokinin concentrations ( Tachibana et al., 1997). Preliminary
xylem sap analyses showed that 20[A-RZT plants had
half the xylem cytokinin concentration of 20-RZT plants
(data not shown). However, variation of cytokinin concentrations and fluxes within a physiologically relevant
range had no effect on stomata in transpiration bioassays
(IC Dodd, CA Beveridge, A Fletcher, and RE Munns,
unpublished observations).
Since no strong evidence was found that alterations in
chemical signalling were responsible for stomatal closure
of 20[A-RZT plants, a role for hydraulic factors was
Growth, stomatal conductance and root temperature 247
reconsidered. One difficulty of working in greenhouses is
that plant water status can fluctuate rapidly with changes
in incident light intensity due to shadows caused by the
greenhouse structure and the passing of clouds. Since the
roots of aeroponically grown plants are continuously
supplied with water, changes in plant water status may
well be transient and limited to periods of high VPD.
This may explain why significant (P<0.10) differences in
Y
between 20-RZT and 20[A-RZT plants were not
shoot
detected ( Table 2).
Conclusions
The data from this study show that the factor(s) controlling leaf initiation and development were independent of
the factor(s) controlling stomatal conductance ( g ) in
s
aeroponically grown Capsicum at different root temperatures. Plants grown at 20-RZT developed faster than
those at A-RZT, despite lower stomatal conductance
under conditions of high VPD (which are characteristic
of sunny days in Singapore greenhouses). Although
hydraulic factors ( low root hydraulic conductivity causing
shoot wilting) are probably responsible for the lower
stomatal conductance of 20-RZT plants, more work is
needed to measure shoot turgor directly, and to understand the factor(s) (presumably chemical signals from the
roots) that regulate leaf initiation and development.
In the RZT transfer experiments, A[20-RZT plants
showed accelerated development while 20[A-RZT plants
showed retarded development, as expected from studies
of plants grown at one RZT. However, stomatal responses
of these plants were the opposite of that expected, in that
A[20-RZT plants showed increased stomatal conductance while 20[A-RZT plants showed stomatal closure.
Root temperature effects on plant development and root
morphology are likely to have modified root hydraulic
conductivity (Lp) independently of any effect of temperature on water viscosity. Alterations in Lp have in turn
affected Y
, which is hypothesized to have directly
shoot
affected the stomata. Confirmation of this hypothesis will
require measurement of Lp in similar RZT transfer experiments, and include manipulation of shoot turgor via root
pressurisation to determine whether Y
can directly
shoot
affect the stomata.
Acknowledgements
We thank the Singapore Department of Environment for
provision of meteorological data, and Dr Brian Loveys, CSIRO
Division of Plant Industry, Adelaide, for the kind gift of [2H ]
6
ABA and independent assays of [ X-ABA]. Mr Andrew Fletcher
is thanked for assistance in optimizing the nitrate assay.
References
Barthes L, Deleens E, Bousser A, Hoarau J, Prioul JL. 1996.
Nitrate use and xylem exudation in detopped wheat seedlings:
an early diagnosis for predicting varietal differences in
nitrogen utilisation? Australian Journal of Plant Physiology
23, 33–44.
Bassirirad H, Radin JW, Matsuda K. 1991. Temperaturedependent water and ion transport properties of barley and
sorghum roots. I. Relationship to leaf growth. Plant
Physiology 97, 426–432.
Behboudian MH, Graves WR, Walsh CS, Korcak RF. 1994.
Water relations, mineral nutrition, growth and 13C discrimination in two apple cultivars under daily episodes of high rootmedium temperature. Plant and Soil 162, 125–133.
Bunce JA. 1977. Leaf elongation in relation to leaf water
potential in soybean. Journal of Experimental Botany 28,
156–161.
Dodd IC, Stikic R, Davies WJ. 1996. Chemical regulation of
gas exchange and growth of plants in drying soil in the field.
Journal of Experimental Botany 47, 1475–1490.
Douglas JS. 1982. Advanced guide to hydroponics. Natraj
Publishers, Dehra Dun, India.
Fiscus EL, Markhart III AH. 1979. Relationship between root
system water transport properties and plant size in Phaseolus.
Plant Physiology 64, 770–773.
Ford YY, Harrison-Murray RS. 1997. Water transport through
ornamental shrub cuttings. Journal of Horticultural Science
72, 687–696.
Gossellin A, Trudel MJ. 1986. Root-zone temperature effects
on pepper. Journal of the American Society for Horticultural
Science 111, 220–224.
Graves WR, Aiello AS. 1997. High root-zone temperature
causes similar changes in water relations and growth of silver
maples from 33 degrees and 44 degrees N latitude. Journal
of the American Society for Horticultural Science 122, 195–199.
Graves WR, Joly RJ, Dana MN. 1991. Water use and growth
of honey locust and tree-of-heaven at high root-zone
temperatures. HortScience 26, 1309–1312.
He J, Lee SK. 1998a. Growth and photosynthetic responses of
three aeroponically grown lettuce cultivars (Lactuca sativa
L.) to different root-zone temperatures and growth irradiances
under tropical aerial conditions. Journal of Horticultural
Science and Biotechnology 73, 173–180.
He J, Lee SK. 1998b. Growth and photosynthetic characteristics
of lettuce (Lactuca sativa L.) under fluctuating hot ambient
temperatures with the manipulation of cool root-zone temperature. Journal of Plant Physiology 152, 387–391.
Jeschke WD, Pate JS. 1992. Temporal patterns of uptake, flow
and utilization of nitrate, reduced nitrogen and carbon in a
leaf of salt-treated castor bean (Ricinus communis L.). Journal
of Experimental Botany 43, 393–402.
Lee SK, Cheong SC. 1996. Inducing head formation of iceberg
lettuce (Lactuca sativa L.) in the tropics through root-zone
temperature control. Tropical Agriculture 73, 34–42.
Markhart III AH, Fiscus EL, Naylor AW, Kramer PJ. 1979.
Effect of temperature on water and ion transport properties
in soybean and broccoli root systems. Plant Physiology
64, 83–87.
McWilliam JR, Kramer PJ, Musser RL. 1982. Temperatureinduced water stress in chilling sensitive plants. Australian
Journal of Plant Physiology 9, 343–352.
Menzel CM, Turner DW, Doogan VJ, Simpson PR. 1994. Root–
shoot interactions in passionfruit (Passiflora sp.) under the
influence of changing root volumes and soil temperatures.
Journal of Horticultural Science 69, 353–364.
248 Dodd et al.
Mercado JA, Reid MS, Valpuesta V, Quesada MA. 1997.
Metabolic changes and susceptibility to chilling stress in
Capsicum annuum plants grown at suboptimal temperature.
Australian Journal of Plant Physiology 24, 759–767.
Newman SE, Davies FT. 1988. High root-zone temperatures,
mycorrhizal fungi, water relations, and root hydraulic
conductivity of container-grown woody plants. Journal of the
American Society for Horticultural Science 113, 138–146.
Olsen JK, Lyons DJ. 1994. Petiole sap nitrate is better than
total nitrogen in dried leaf for indicating nitrogen status and
yield responsiveness of Capsicum in subtropical Australia.
Australian Journal of Experimental Agriculture 34, 835–843.
Palmer SJ, Berridge DM, McDonald AJS, Davies WJ. 1996.
Control of leaf expansion in sunflower (Helianthus annuus L.)
by nitrogen nutrition. Journal of Experimental Botany 47,
359–368.
Schurr U, Schulze ED. 1996. Effects of drought on nutrient and
ABA transport in Ricinus communis. Plant, Cell and
Environment 19, 665–674.
Shashidhar VR, Prasad TG, Sudharshan L. 1996. Hormone
signals from roots to shoots of sunflower (Helianthus annuus
L.): moderate soil drying increases delivery of abscisic acid
and depresses delivery of cytokinins in xylem sap. Annals of
Botany 78, 151–155.
Sheppard SC, Racz GJ, Martin SL. 1986. Critical concentration
of tissue phosphorus when root temperature depresses growth
rate. Journal of Experimental Botany 37, 1265–1273.
Tachibana S, Du YC, Wang YuH, Kitamura F. 1997. Implication
of endogenous cytokinins in the growth inhibition of
cucumber plants by supraoptimal root-zone temperature.
Journal of the Japanese Society for Horticultural Science 66,
549–555.
Trejo CL, Clephan AL, Davies WJ. 1995. How do stomata read
abscisic acid signals? Plant Physiology 109, 803–811.
Turner NC. 1974. Stomatal behaviour and water status of
maize, sorghum and tobacco under field conditions. Plant
Physiology 53, 360–365.
Udagawa Y, Ito T, Gomi K. 1991. Effects of root temperature
on the absorption of water and mineral nutrients by
strawberry plants ‘cultivar Reiko’ grown hydroponically.
Journal of the Japanese Society for Horticultural Science 59,
711–718.
Wilkinson S, Corlett JE, Oger L, Davies WJ. 1998. Effects of
xylem pH on transpiration from wild-type and flacca tomato
leaves. Plant Physiology 117, 703–709.
Zhang J, Davies WJ. 1989. Sequential response of whole plant
water relations to prolonged soil drying and the involvement
of xylem sap ABA in the regulation of stomatal behaviour
of sunflower plants. The New Phytologist 113, 167–174.