Mycorrhizal symbiosis and response of sorghum plants to combined drought and salinity stresses

Abstract

Arbuscular mycorrhizal (AM) symbiosis can confer increased host resistance to drought stress, although the effect is unpredictable. Since AM symbiosis also frequently increases host resistance to salinity stress, and since drought and salinity stress are often linked in drying soils, we speculated that the AM influence on plant drought response may be partially the result of AM influence on salinity stress. We tested the hypothesis that AM-induced effects on drought responses would be more pronounced when plants of comparable size are exposed to drought in salinized soils. In two greenhouse experiments, several water relations characteristics were measured in sorghum plants colonized by Glomus intraradices (Gi), Gigaspora margarita (Gm) or a mixture of AM species, during a sustained drought following exposure to salinity treatments (NaCl stress, osmotic stress via concentrated macronutrients, or soil leaching). The presence of excess salt in soils widened the difference in drought responses between AM and nonAM plants in just two instances. Days required for plants to reach stomatal closure were similar for Gi and nonAM plants exposed to drought alone, but with exposure to combined NaCl and drought stress, stomates of Gi plants remained open 17-22% longer than in nonAM plants. Promotion of stomatal conductance by Gm occurred with exposure to NaCl/drought stress but not with drought alone or with soil leaching before drought. In other instances, however, the addition of salt tended to nullify an AM-induced change in drought response. Our findings confirm that AM fungi can alter host response to drought but do not lend much support to the idea that AM-induced salt resistance might help explain why AM plants can be more resilient to drought stress than their nonAM counterparts.

Full text

Journal of Plant Physiology 163 (2006) 517—528
Mycorrhizal symbiosis and response of sorghum
plants to combined drought and salinity stresses
Keunho Cho
a
, Heather Toler
a
, Jaehoon Lee
b
, Bonnie Ownley
c
,
Jean C. Stutz
d
, Jennifer L. Moore
a
, Robert M. Auge´
a,

a
Department of Plant Sciences, University of Tennessee, 2431 Joe Johnson Drive, Knoxville, TN 37996-4561, USA
b
Department of Biosystems Engineering and Environmental Science, University of Tennessee, Knoxville, TN 37996-
4561, USA
c
Department of Entomology and Plant Pathology, University of Tennessee, Knoxville, TN 37996-4561, USA
d
Department of Plant Biology, Arizona State University, Tempe, AZ 85287-1601, USA
Received 27 January 2005; accepted 2 May 2005
Summary
Arbuscular mycorrhizal (AM) symbiosis can confer increased host resistance to drought
stress, although the effect is unpredictable. Since AM symbiosis also frequently
increases host resistance to salinity stress, and since drought and salinity stress are
often linked in drying soils, we speculated that the AM influence on plant drought
response may be partially the result of AM influence on salinity stress. We tested the
hypothesis that AM-induced effects on drought responses would be more pronounced
when plants of comparable size are exposed to drought in salinized soils. In two
greenhouse experiments, several water relations characteristics were measured in
sorghum plants colonized by Glomus intraradices (Gi), Gigaspora margarita (Gm)ora
mixture of AM species, during a sustained drought following exposure to salinity
treatments (NaCl stress, osmotic stress via concentrated macronutrients, or soil
leaching). The presence of excess salt in soils widened the difference in drought
responses between AM and nonAM plants in just two instances. Days required for
plants to reach stomatal closure were similar for Gi and nonAM plants exposed to
drought alone, but with exposure to combined NaCl and drought stress, stomates of Gi
plants remained open 17–22% longer than in nonAM plants. Promotion of stomatal
conductance by Gm occurred with exposure to NaCl/drought stress but not with
drought alone or with soil leaching before drought. In other instances, however, the
addition of salt tended to nullify an AM-induced change in drought response. Our
ARTICLE IN PRESS
www.elsevier.de/jplph
KEYWORDS
Arbuscular
mycorrhizal
symbiosis;
Drought stress;
Lethal water
potential;
Salinity stress;
Sorghum;
Stomatal
conductance
0176-1617/$ - see front matter &2005 Elsevier GmbH. All rights reserved.
doi:10.1016/j.jplph.2005.05.003
Abbreviations: AM arbuscular mycorrhizal; AZ species assemblage of AM fungi from a semi-arid grassland in Arizona; Gi Glomus
intraradices;Gm Glomus margarita;C
p
osmotic potential; C100
posmotic potential at full turgor; g
s
stomatal conductance; Cwater
potential

Corresponding author. Tel.: +865 974 7324; fax: +865 974 1947.
E-mail address: auge@utk.edu (R.M. Auge´).
URL: http://plantsciences.utk.edu/auge.htm.

findings confirm that AM fungi can alter host response to drought but do not lend much
support to the idea that AM-induced salt resistance might help explain why AM plants
can be more resilient to drought stress than their nonAM counterparts.
&2005 Elsevier GmbH. All rights reserved.
Introduction
Arbuscular mycorrhizal (AM) symbiosis is often
alleged to improve plant resistance to drought
stress. Several studies have demonstrated this
under varying experimental conditions (e.g. Sub-
ramanian and Charest, 1998;Ruiz-Lozano and
Azco´n, 2000;Porcel et al., 2003), while others
revealed little or no AM enhancement of resistance
(e.g. Hetrick et al., 1987;Simpson and Daft, 1991).
There are reports of AM-induced increases in
physiological drought tolerance, involving both
increased dehydration avoidance and dehydration
tolerance (Allen and Boosalis, 1983;Davies et al.,
1993). Most experiments examining AM effects on
drought resistance have shown that when the
symbiosis improves host drought resistance, it does
so by aiding drought avoidance (Auge´, 2001). The
AM influence on plants in drying soils remains
unpredictable and uncertain, particularly in soils
with adequate phosphorus.
AM symbiosis has frequently increased resilience
of host plants to salinity stress, perhaps with
greater consistency than to drought stress.
Growth in saline soils was increased by inoculation
with Glomus spp, with AM plants having
increased phosphate and decreased Na concentra-
tions in shoots compared to uninoculated
controls (Pfeiffer and Bloss, 1987;Giri and Mukerji,
2004). Salt resistance was improved by AM
colonization in maize (Feng et al., 2002), mung
bean (Jindal et al., 1993) and clover (Ben Khaled
et al., 2003), with the AM effect correlated
with improved osmoregulation or proline accumu-
lation. AM colonization also improved NaCl
resistance in tomato, with extent of improvement
related to salt sensitivity of the cultivar (Al-Karaki,
2000;Al-Karaki et al., 2001). AM improvement
of salt resistance has usually been associated
with AM-induced increases in P acquisition and
plant growth, although two of three AM fungi
tested were able to protect cucumber plants from
NaCl stress compared to similarly sized nonAM
plants (Rosendahl and Rosendahl, 1991). Alfalfa
was also more effectively protected against salinity
stress by AM symbiosis than by P supplementation
(Azco´n and El-Atrash, 1997), and the improvement
of NaCl resistance in lettuce induced by several AM
fungi was not attributed to nutrition (Ruiz-Lozano
et al., 1996).
Since solutes can concentrate in the soil solution
just outside roots as soil dries (Stirzaker and
Passioura, 1996), and since AM symbiosis often
increases plant resistance to salinity stress, we
speculated that the amount of salt in drying soil
may be one experimental factor that could explain
why AM fungi increased drought resistance in some
studies but not in others, i.e. perhaps AM effects on
drought resistance are linked to AM effects on salt
resistance; in those reports where AM symbiosis did
improve drought resistance, AM fungi may have
helped to overcome plant susceptibility to an
osmotic or NaCl stress that developed as soil dried.
Osmotic potential (C
p
) and NaCl concentration of
soil are typically not reported in drought studies, so
it is difficult to examine the mycorrhizal literature
for such trends.
We conducted experiments to test whether AM
effects on physiological drought resistance would
be more pronounced under saline conditions. We
exposed sorghum to various salinity treatments just
prior to exposing them to drought. Plants must
often endure both drought and salinity stress in arid
and semi-arid regions, and it would be useful to
know if AM symbiosis can increase resistance to
these combined stresses.
Materials and methods
Experiment 1: Glomus intraradices, NaCl
and osmotic stress
Plant materials and culture
One hundred and five 2.8-L plastic pots were
seeded with Sorghum bicolor L. cv Dekalb DK40Y on
6 February 2002, with 5 seeds per pot. Potting
medium was autoclaved silica sand (commercial
medium grade, No. 1962-51, Quikrete, Atlanta, GA,
USA). Thirty-five pots received pot culture colo-
nized by Glomus intraradices Schenck and Smith
INVAM isolate UT143 (Gi) and 70 pots received
nonAM pot culture. AM and nonAM pot cultures
were established on sorghum in a loam:sand (1:1)
mixture and were 9 months old at the time of
inoculation. 100–150 g of either AM or nonAM pot
culture were banded beneath seeds. In addition to
using nonAM pot cultures grown under similar
conditions as AM pot culture, similar soil microbial
ARTICLE IN PRESS
K. Cho et al.518

populations were encouraged in AM and nonAM pots
by also applying water filtrates of AM inoculum to
nonAM pots (44-mm sieve).
Plants were fertilized weekly with water soluble
fertilizer at 5.6 mM N (Peters, N:P:K ¼15:0:5,
Scotts-Sierra Horticultural Product Co., Marysville,
OH, USA) and biweekly with a micronutrient
solution at 0.02 mM Fe (Microplex, Miller Chemical
and Fertilizer Co., Hanover, PA, USA). Phosphorus
was supplied once per week as 0.8 mM KH
2
PO
4
to
AM plants and one group of nonAM plants and as
1.6 mM KH
2
PO
4
to a second group of nonAM plants.
Plants were grown in a glasshouse in Knoxville, TN,
with temperature maintained at 25–29 1C/18–23 1C
(day/night) under natural light. The glasshouse was
covered between May and October with 55% shade
cloth to aid in temperature control. Plants were
watered as needed prior to stress applications.
The AM and nonAM soil treatments were allowed
to establish for 10 months before initiating the
stress treatments, following the procedures of
Auge´ et al., 2004b. To renew shoot growth during
the soil establishment phase, plants were sheared
three times (1 May, 17 June, 16 October 2002) and
pots reseeded twice (16 August, 12 November 2002;
5 seeds per pot) during the 10-month establishment
period. High-pressure sodium lamps (400 W) were
installed 80 cm above the pots (16 h day
1
)to
supplement low natural light during the winter
months.
Leaves from the final shearing and crowns were
collected for each pot and dry weights determined,
to assess which nonAM treatment group was closest
in size to the AM plants. AM and nonAM pots given
low P were selected for further study because they
had similar shoot dry weight (4.6870.16 and
4.6570.11 g pot
1
, respectively).
Drought and salinity treatments
When plants in the final planting were 10 weeks
old, and tests confirmed that AM plants were
colonized and nonAM plants were free of coloniza-
tion, stress treatments were applied. Soils were
exposed to three salinity stress treatments just
prior to initiating the drought treatment: a
concentrated macronutrient solution, a NaCl solu-
tion or distilled water. The drought treatment
consisted of allowing plants to dry, with the last
day that plants were watered designated as ‘‘day
0’’ (27 January 2003).
The first salinity stress applications were applied
7 d before withholding water (day-7). A macronu-
trient solution was used to lower soil C
p
and induce
osmotic stress. It was composed of 40 mM MgSO
4
,
90 mM Ca(NO
3
)
2
, 1.6 mM KH
2
PO
4
, 62 mM KNO
3
and
19 mM NH
4
NO
3
(Auge´ et al., 1992). The solution was
adjusted to a water potential (C)of0.4 MPa for
the first application on day 7 and 0.8 MPa for the
second application on day 0. Plants were watered
with tap water once (day 4) between the first and
second applications of the salt solutions. The
second salinity stress treatment was NaCl solutions,
applied at 40 mM for the first application on day 7
and 80 mM for the second application on day 0. Cof
the 40 mM and 80 mM NaCl solutions were 0.19
and 0.37 MPa, respectively. The third salinity
stress treatment was a distilled water control. For
each application, 200 mL of distilled water, NaCl or
macronutrient solution was applied to pots. Water
was withheld from all pots after the second
application. The more dilute first application was
applied 1 week prior to the full application,
followed by one irrigation 3 d later with tap water,
to ease the plants into the salinity stress.
Plant, soil and solution measurements
Cof solutions was measured with a chilled mirror
dewpoint hygrometer (WP4, Decagon Devices Inc.,
Pullman WA, USA) calibrated with NaCl solutions.
Soil electrical conductivity (EC) of each pot was
measured with an EC meter (AR 20, Fisher Scientific
Inc., Pittsburgh, PA, USA) after calibration (0.01 N
KCl solution) by the 1:5 soil/water suspension
method (Rayment and Higginson, 1992).
Beginning on day 3 of the drought stress period,
stomatal conductance (g
s
) of plants in each pot was
measured each day at 1300 EST on abaxial leaf
surfaces with an automatic-cycling porometer
(AP4, Delta-T Devices, Cambridge, England). Three
leaves per pot were measured each day, at the
distil end of the largest, unshaded leaves. Daily
measurements continued until the average g
s
for
the foliage in a particular pot declined to below
10 mmol m
2
s
1
. This point was defined as the
stomatal closure point, as stomates of sorghum
leaves tend not to close completely. Stomatal
conductance was also measured 2–3 times per
week prior to application of salt treatment solu-
tions.
Lethal leaf Cwas measured with thermocouple
psychrometers (TruPsi, Decagon Devices Inc.) as
described before (Auge´ et al., 2001a, b). Sampling
was performed between 1030 and 1130 h EST.
For measurement of initial (pre-stress) C
p
, one
leaf from each pot in each treatment was excised
on day-7. The leaves were cut into two halves. The
proximal half was immediately sealed in a syringe
and plunged into liquid N
2
for measurement of C
p
using a vapor pressure osmometer (model 5500XR,
Wescor Inc., Logan UT, USA). The base of the distal
half was submerged in distilled water in a covered
beaker and rehydrated in a refrigerator (4 1C)
ARTICLE IN PRESS
Mycorrhizal sorghum, salinity and drought stress 519

overnight for measurement of C
p
at full turgor
ðC100
pÞas described before (Chapman and Auge´,
1994).
For measurements of soil C, samples (about
2.5 mL) were excavated from the middle (14–17 cm
from top) and bottom (27–30 cm from top) of each
pot at each time period, from flaps cut in the pot,
immediately sealed in sample cups and Cmeasured
using the WP4 hygrometer. Flaps in pots were
sealed with duct tape following sampling.
Leaf C, leaf C
p
, leaf C100
pand soil Cwere
measured at the stomatal closure and lethal points
as well as 1 d after application of the half- and full-
strength soil solutions.
On day-10, shoot dry weight was determined for
plants in five extra AM and nonAM pots. Soil from
each pot was mixed and subsampled for determina-
tion of soil hyphal density, root colonization and
root density. Soil hyphal density was measured as
described before (Auge´ et al., 2001b), on 10 g
subsamples. Roots were carefully excavated from
another 25 g subsample of each replicate for
measurement of root length using scanning equip-
ment and imaging software (WinRhizo, Regent
Instruments Inc., Quebec City, Canada). Root
colonization was quantified on 100 0.5-cm root
pieces from each pot, after clearing with 10% KOH
in an autoclave at 121 1C for 15 min, staining with
trypan blue for 1 h, and destaining.
On day-10, leaf P concentration of one of the
largest, most recently expanded leaves of four
plants from each treatment was determined spec-
trophotometrically using the vanadate-molybdate-
yellow method on samples dry-ashed with magne-
sium nitrate at 700 1C for 2 h and digested in nitric
acid (Chapman and Pratt, 1961).
Experiment 2: Gigaspora margarita, a semi-
arid AM mix, and NaCl stress
Plant materials and culture
One-hundred and eight 2.8-L plastic pots were
seeded with S. bicolor L. cv Dekalb DK40Y on 18
April 2003, with 8 seeds per pot. The potting
medium was composed of autoclaved silica sand as
in Experiment 1. Thirty-six pots received pot
culture colonized by Gigaspora margarita Gerde-
mann & Trappe INVAM isolate 215 (Gm), 36 pots
received pot culture colonized by a mixture of AM
species originally isolated from a semiarid grassland
in Arizona (AZ mix, described in Auge´ et al., 2003),
and 36 pots received nonAM pot culture. Gm and
the AZ-mix pot cultures were established on
sorghum in a loam:sand (1:1) mixture and were 12
months old at the time of inoculation. Composition
of the AZ-mix greenhouse pot culture in terms of
spore abundance was 88% Gl. intraradices, 10%
Glomus AZ 123, 2% Acaulospora rehmii.
Other cultural details were as described for
Experiment 1. Phosphorus was supplied once a
week as 0.8 mM KH
2
PO
4
to plants in each AM
treatment and to nonAM plants. After emergence,
pots were thinned back to the six strongest
seedlings.
Drought and salinity treatments; plant, soil
and solution measurements
Drought and salinity treatments were applied as
described for Experiment 1, with one modification
of the salinity treatments: the macronutrient
salinity treatment was replaced by a leaching
treatment. In this treatment, pots were leached
with distilled water (approximately 600 mL applied
to each pot in three 200-mL leachings) to remove
some soil solutes. The first, half-strength applica-
tion of soil solutions occurred 11 June 2003 (day-7)
and the second, full-strength application 18 June
2003 (day 0). Plant, soil and solution measurements
were made as in Experiment 1, with two modifica-
tions: g
s
measurements began on day 1, and
measurements were not made at the lethal point.
Experimental design and statistical analysis
For each factorial experiment, pots were ar-
ranged in a completely randomized block design.
Experiment 1 (2 3 factorial) had two mycorrhizal
treatments and Experiment 2 (3 3 factorial) had
three mycorrhizal treatments. Each experiment
had three salinity treatments: an NaCl solution,
concentrated macronutrient solution and water for
Experiment 1; and an NaCl solution, water and
water leaching for Experiment 2. Each level of each
factor had six replicates for each experiment.
ANOVA was performed using the General Linear
Model procedure (SAS, Cary, NC, USA). Duncan’s
mean separation tests and ttests were used to
examine differences among means.
Results
Experiment 1: Glomus intraradices, NaCl
and osmotic stress
Shoot, root, fungal and soil EC characteristics
NonAM and Gi plants had similar shoot dry
weights, leaf [P], and root mass and length
densities (Table 1). Roots of Gi plants developed
considerable AM colonization. NonAM plants re-
ARTICLE IN PRESS
K. Cho et al.520

mained nonmycorrhizal. Soil in Gi pots had about
seven times more hyphal length than soil in nonAM
pots.
Soil EC averaged 0.20 dS m
1
in each of the six
treatment groups 2 d prior to application of the first
saline treatment. Application of the half-strength
solutions on day-7 caused soil EC to increase about
3-fold for plants given NaCl and about 6- to 8-fold in
plants given the macronutrient solution, relative to
control plants given distilled water. Application of
the full-strength solutions on day 0 caused soil EC
to increase about 4- to 5-fold for plants given NaCl
and about 20- to 30-fold in plants given the
macronutrient solution, relative to control plants
given distilled water.
Plant water relations
Stomatal conductance was higher in Gi than in
nonAM plants before drought was initiated (Fig. 1).
Averaged over all replicates of all treatments on all
measurement days before day 0, mean pre-drought
g
s
of Gi and nonAM plants was 265 mmol m
2
s
1
and
213 mmol m
2
s
1
, respectively (significantly differ-
ent, P¼0:02). Average g
s
of Gi plants during the
drying period was higher than that of nonAM plants,
by 32% with drought stress alone (P¼0:03), by 38%
with NaCl/drought stress (P¼0:007), and by 51%
with osmotic/drought stress (P¼0:006) (Fig. 2).
Plants in individual pots reached stomatal closure
between days 9 and 12.
AM symbiosis did not affect leaf C1 d after
exposure to the half-strength (first application) and
full-strength (second application) solutions in any
of the three salinity treatments (Table 2). Leaf Cat
stomatal closure was 1.2 MPa higher in Gi than in
nonAM plants exposed to drought stress alone. Leaf
Cat stomatal closure was similar in Gi and nonAM
plants exposed to NaCl/drought stress and to
osmotic/drought stress. Leaf Cat the lethal point
was similar in Gi and nonAM plants exposed to each
of the soil solutions. Plants in individual pots
reached the lethal point between days 14 and 25.
AM symbiosis did not affect leaf C
p
in any of the
three salinity treatments 1 d after exposure to the
half-strength and full-strength solutions or at
stomatal closure (Table 3). Leaf C
p
at the lethal
point was 0.70 MPa higher in Gi than in nonAM
plants exposed to osmotic/drought stress. Leaf C
p
at the lethal point was similar in Gi and nonAM
plants exposed to NaCl/drought stress.
Leaf C100
pwere measured 1 d after application of
the first and second salinity treatment solutions,
and at the lethal point. AM symbiosis had no effect
on leaf C100
pon any of these days nor on osmotic
adjustment (data not shown).
NonAM plants reached stomatal closure 2 d soon-
er than Gi plants when exposed to drought in the
absence of salinity (Table 4). With exposure to
either the NaCl or the macronutrient solutions
during drought, nonAM and Gi plants reached
stomatal closure at the same time. Plants remained
alive longer when droughted with exposure to
either salt solution relative to plants exposed to
drought alone: 18–23 d vs. 14–15 d. NonAM and Gi
ARTICLE IN PRESS
Table 1. Shoot dry weight, leaf [P], and root and fungal characteristics of representative sorghum plants of each AM
treatment on day-10 in Experiments 1 and 2
Nonmycorrhizal Gl. intraradices
Experiment 1
Shoot dry weight (g pot
1
) 4.5 a 4.4 a
Leaf [P] (mg g
1
DW) 1.5 a 1.5 a
Root mass density (mg g
1
dry soil) 1.3 a 1.4 a
Root length density (cm g
1
dry soil) 13.8 a 15.6 a
AM root colonization (%) 0 a 53 b
Soil hyphal density (m g
1
dry soil) 0.2 a 1.2 b
Experiment 2 Nonmycorrhizal AZ mix Gi. margarita
Shoot dry weight (g pot
1
) 1.2 b 0.8 a 1.4 b
Leaf [P] (mg g
1
DW) 1.5 a 1.8 b 1.8 ab
Root mass density (mg g
1
dry soil) 0.5 a 0.3 a 0.6 a
Root length density (cm g
1
dry soil) 6.1 a 6.0 a 8.6 b
AM root colonization (%) 0 a 36 c 24 b
Soil hyphal density (m g
1
dry soil) 0.0 a 0.4 c 0.2 b
Values for AM root colonization represent total colonization levels and indicate presence of hyphae, arbuscules and/or vesicles. n¼5
for each parameter except leaf [P], where n¼4. Within rows, means followed by different letters are significantly different (Pp0:05).
Mycorrhizal sorghum, salinity and drought stress 521

plants reached the lethal point in the same amount
of time when droughted without added salt as well
as when droughted with exposure to the macro-
nutrient solution. NonAM plants died more quickly
than Gi plants when exposed to drought and NaCl.
Soil water relations
After the first half-strength application of soil
solutions, soil Cwas 0.05 MPa lower in Gi than in
nonAM plants exposed to osmotic stress (Table 5).
Soil Cwas similar in Gi and nonAM plants exposed
to NaCl after the first and second applications, as
well as to osmotic stress after the second applica-
tion.
Soil Cat the point of stomatal closure was
substantially higher in Gi than in nonAM plants
exposed to drought alone, by 2.36 MPa (Table 4).
Soil Cwas appreciably higher in Gi than in nonAM
plants at the lethal point with exposure to NaCl/
drought stress (by 1.25 MPa) and to osmotic/
drought stress (by 1.77 MPa).
Experiment 2: Gigaspora margarita, a semi-
arid AM mix, and NaCl stress
Shoot, root, fungal and soil EC characteristics
Gm and nonAM plants had similar shoot dry
weights and root length densities (Table 1). Shoots
of AZ-mix plants were markedly smaller than
nonAM and Gm plants. NonAM and Gm plants had
similar leaf [P], and AZ-mix plants had slightly
higher leaf [P] than nonAM plants. The smaller AZ-
mix plants had about 60% of the root mass density
of nonAM plants. Plants examined in Experiment 2
were younger and smaller than those in Experiment
1(Table 1).
Roots of Gm and AZ-mix plants each developed
considerable AM colonization (Table 1). NonAM
ARTICLE IN PRESS
Figure 1. Stomatal conductance of sorghum plants before and following application of the soil solution treatments, for
Experiment 1 (a–c) and Experiment 2 (d–f). Symbols represents the mean of 18 measurements (three leaves from 6 pots
of each treatment on each day). Standard error of the means for pre-drought g
s
are represented by vertical bar in each
panel. Standard error of the means for g
s
during the drought period were smaller than the height of a symbol.
K. Cho et al.522

plants remained nonmycorrhizal. The younger
plants of Experiment 2 developed less soil hyphae
than in the larger and older Experiment 1 plants.
Despite the smaller shoot and root mass of host
plants, soil in pots with plants colonized by the AZ
mix had greater hyphal length than soil in either
nonAM or Gm pots.
Soil EC was not affected by AM symbiosis after
either the first or second applications of soil
solution treatments.
Plant water relations
Pre-drought g
s
was affected in Experiment 2 by
mycorrhizal symbiosis (Fig. 1). Averaged over all
replicates of all treatments on all measurement
days before day 0, mean pre-drought g
s
of Gm, AZ-
mix and nonAM plants was 296, 205 and
248 mmol m
2
s
1
, respectively, with Gm signifi-
cantly higher than AZ (P¼0:0001) and nonAM
(P¼0:02). Average g
s
during the drying period
was higher in Gm plants than in nonAM plants by
ARTICLE IN PRESS
Figure 2. Average stomatal conductance within treatments during the drying episode for Experiments 1 and 2. n¼162
(Experiment. 1), n¼360 (Experiment. 2). Asterisk denotes that Gi or Gm and nonAM treatments differed significantly
(Pp0:05). Vertical lines represent standard errors.
Mycorrhizal sorghum, salinity and drought stress 523

ARTICLE IN PRESS
Table 2. Leaf Cof nonAM and AM sorghum plants during Experiments 1 and 2
Soil solution treatment Mycorrhizal treatment Leaf C(MPa)
First application Second application Stomatal closure Lethal point
Experiment 1
Drought nonAM 0.66 ab 1.32 a 3.86 b 6.88 b
Gi 0.52 a 1.40 ab 2.67 a 7.03 b
NaCl/drought stress nonAM 0.84 b 1.67 bc 2.50 a 5.71 a
Gi 0.80 b 1.58 abc 2.42 a 4.36 a
Osmotic/drought stress nonAM 0.81 b 1.81 c 2.86 a 4.79 a
Gi 0.83 b 1.76 c 2.93 a 4.02 a
Experiment 2
Drought nonAM 0.94 a 1.00 a 2.26 a
Gm 0.99 a 0.92 a 2.38 a
AZ mix 0.93 a 0.85 a 2.39 a
NaCl/drought stress nonAM 0.93 a 0.96 a 2.48 a
Gm 0.94 a 0.87 a 2.13 a
AZ mix 0.97 a 0.93 a 2.66 a
Leach/drought stress nonAM 0.94 a 0.85 a 2.61 a
Gm 0.94 a 0.84 a 2.46 a
AZ mix 0.91 a 0.80 a 2.03 a
First applications, made at day 7, were a 40 mM NaCl solution (Expt. 1 and 2), distilled water (Expt. 1 and 2), 0.4 MPa macronutrient
solution (Expt. 1), or three applications of distilled water to leach the soil medium (Expt. 2). Second applications, made at day 0 when
drought was initiated, were an 80 mM NaCl solution (Expt. 1 and 2), distilled water (Expt. 1 and 2), 0.8 MPa macronutrient solution
(Expt. 1), or three applications of distilled water to further leach the soil medium (Expt. 2). n¼6. Within columns, means followed by
different letters are significantly different (Pp0:05). Gi ¼Glomus intraradices, Gm ¼Gigaspora margarita, AZ mix ¼a semi-arid
mixture of species.
Table 3. Leaf C
p
of nonAM and AM sorghum plants during Experiments 1 and 2
Soil solution treatment Mycorrhizal treatment Leaf C
p
(MPa)
First application Second application Stomatal closure Lethal point
Experiment 1
Drought nonAM 0.92 a 1.07 a 1.93 a na
Gi 0.90 a 0.98 a 1.85 a na
NaCl/drought stress nonAM 1.10 b 1.47 b 2.06 a 3.22 a
Gi 1.00 ab 1.40 b 1.95 a 3.01 a
Osmotic/drought stress nonAM 0.97 ab 1.65 c 2.66 b 4.62 c
Gi 0.95 ab 1.49 b 2.60 b 3.92 b
Experiment 2
Drought nonAM 0.89 ab 0.85 a 1.53 a
Gm 0.87 ab 0.90 ab 1.63 abc
AZ mix 0.83 a 0.90 ab 1.53 a
NaCl/drought stress nonAM 0.92 b 1.05 c 1.82 bc
Gm 0.83 a 0.89 ab 1.78 bc
AZ mix 0.89 ab 0.91 ab 1.85 c
Leach/drought stress nonAM 0.91 b 0.91 b 1.61 ab
Gm 0.84 a 0.89 ab 1.54 a
AZ mix 0.86 ab 0.93 b 1.49 a
n¼6. ‘‘na’’ indicates leaves were too dehydrated to provide sufficient sap for measurement. Within columns, means followed by
different letters are significantly different (Pp0:05). Gi ¼Glomus intraradices, Gm ¼Gigaspora margarita, AZ mix ¼a semi-arid
mixture of species.
K. Cho et al.524

28% in the NaCl/drought treatment (P¼0:05) and
similar in Gm and nonAM plants in the drought and
leach/drought treatments (P¼0:31 and 0.21,
respectively) (Fig. 2). Plants in individual pots
reached stomatal closure between days 8 and 20.
Leaf Cwas unaffected by AM symbiosis, 1 d after
the first and second applications of the soil
solutions and at stomatal closure (Table 2).
Leaf C
p
was 0.10–0.15 MPa higher in Gm than in
nonAM plants after the first and second applications
of the NaCl solutions and 0.07 MPa higher in Gm
than in nonAM plants after the first soil leaching
treatment (Table 3). Leaf C
p
at the point of
stomatal closure was lowered by about 0.3 MPa in
the NaCl/drought treatment relative to drought in
the absence of salinity in nonAM and AZ-mix plants
but was not affected in Gm plants. Absolute values
of leaf C
p
at stomatal closure were similar among
the three mycorrhizal treatments, with drought
stress, with NaCl/drought stress or with leach/
drought stress. Leaf C100
pwas not affected by AM
symbiosis.
ARTICLE IN PRESS
Table 5. Soil Cof nonAM and AM sorghum plants
Soil solution treatment Mycorrhizal treatment Soil C(MPa)
First application Second application Stomatal closure Lethal point
Experiment 1
Drought nonAM 0.00 a 0.01 a 5.85 c na
Gi 0.00 a 0.05 a 3.49 b na
NaCl/drought stress nonAM 0.00 a 0.47 b 1.88 a 5.20 b
Gi 0.03 a 0.37 b 1.38 a 3.95 a
Osmotic/drought stress nonAM 0.26 b 1.00 c 2.08 a 5.69 b
Gi 0.31 c 0.87 c 1.87 a 3.92 a
Experiment 2
Drought nonAM 0.00 a 0.01 a 2.74 abc
Gm 0.12 a 0.02 a 4.78 d
AZ mix 0.05 a 0.05 a 3.62 cd
NaCl/drought stress nonAM 0.12 a 0.21 b 1.40 a
Gm 0.07 a 0.08 ab 1.67 a
AZ mix 0.08 a 0.19 b 1.91 ab
Leach/drought stress nonAM 0.06 a 0.01 a 3.56 cd
Gm 0.02 a 0.00 a 3.90 cd
AZ mix 0.01 a 0.09 a 3.06 bc
n¼12 (two subsamples, middle and bottom of pot, from six pots of each treatment).‘‘na’’ for droughted plants at lethal point in
Experiment 1 signifies soils were dry beyond the range of measurement capability of the instrument (o40 MPa). Within columns,
means followed by different letters are significantly different (Pp0:05). Gi ¼Glomus intraradices, Gm ¼Gigaspora margarita,AZ
mix ¼a semi-arid mixture of species.
Table 4. Experiment 1. Days required for plants to reach stomatal closure and the lethal point
Soil solution treatment Mycorrhizal treatment Days to:
Stomatal closure Lethal point
Drought nonAM 9.3 a 14.0 a
Gl. intraradices 11.2 cd 15.0 a
NaCl/drought stress nonAM 11.3 cd 17.8 b
Gl. intraradices 11.8 d 20.0 c
Osmotic/drought stress nonAM 10.0 ab 23.3 d
Gl. intraradices 10.5 bc 22.8 d
n¼6. Within columns, means followed by different letters are significantly different (Pp0:05).
Mycorrhizal sorghum, salinity and drought stress 525

Soil water relations
Soil Cwas not affected by AM symbiosis after the
first and second applications of the soil solution
treatments (Table 5). With drought stress, soil Cat
stomatal closure was 2 MPa lower for Gm than for
nonAM plants. Soil Cat stomatal closure was
unaffected by AM symbiosis in the NaCl/drought
or leach/drought treatments. Average soil Cat
stomatal closure across soil solution treatments
was 3.45 MPa for Gm plants and 2.57 MPa for
nonAM plants (P¼0:03).
Discussion
Our objective was to test two hypotheses, with a
successful demonstration of the first a prerequisite
for testing the second: (1) AM symbiosis affects
plant drought response relative to nonAM plants of
comparable size, and (2) the AM-induced effect on
drought response is more pronounced in plants
when they are exposed to drought in salinized soils.
Because AZ-mix plants were considerably smaller
than nonAM plants in Experiment 2, they are not
considered further here in tests of the hypotheses.
The discussion focuses on comparisons of AM and
nonAM plants of similar size: Gi and nonAM plants in
Experiment 1, and Gm and nonAM plants in
Experiment 2. Plant size, regardless of mycorrhizal
status of roots, can affect g
s
and other plant water
relations responses to drought (e.g. Ebel et al.,
1996).
Hypothesis 1. Viewed as a whole, the mycorrhizal
literature indicates that AM symbiosis tends to have
more effect on leaf g
s
than on leaf Cor C
p
(Auge´,
2001). We also observed this in these experiments.
Before drought was initiated, colonization by Gi in
Experiment 1 and by Gm in Experiment 2 resulted
in promotion of g
s
by 24% and 19%, respectively,
averaged over soil solution treatments on all pre-
drought days. This is within the range of promotion
of g
s
by Gi and Gm in another experiment with
adequately watered sorghum (Auge´ et al., 2004a).
Integrating the AM effect on g
s
over all soil solution
treatments and days of drought, Gi increased g
s
by
39% and Gm increased g
s
by 17%.
AM effects on leaf C, leaf C
p
and soil Cat
stomatal closure or at the lethal point were more
sporadic than those on g
s
. Still, several Gi effects
on these water relations parameters were ob-
served. Hypothesis 1 was true for the following
parameters and situations for Gi and nonAM plants:
leaf Cat stomatal closure of plants exposed to
drought stress alone; leaf C
p
at the lethal point of
plants exposed to osmotic/drought stress; soil Cat
stomatal closure of plants exposed to drought
stress alone; soil Cat the lethal point of plants
exposed to NaCl/drought or to osmotic/drought
stress; days of drying required for plants exposed to
drought alone to reach stomatal closure; days of
drying required for plants exposed to NaCl/drought
to reach the lethal point. Gm had less effect than
Gi on host water relations responses to drought.
Hypothesis 1 was true in two instances for Gm and
nonAM plants. Soil Cat stomatal closure was
markedly lower in Gm vs. nonAM plants. And
relative to drought in the absence of salinity,
exposure to NaCl during drought resulted in a
significant decline in leaf C
p
at the point of
stomatal closure in nonAM plants but not in Gm
plants.
Hypothesis 2. Our second hypothesis, that the AM-
induced effect on host plants during drought would
be more pronounced when plants were exposed to
soil drying in salinized soils, was true for just two
instances. Days required for plants to reach
stomatal closure were similar for Gi and nonAM
plants exposed to drought alone, but Gi plants kept
stomates open 2 d longer than nonAM plants with
exposure to combined NaCl/drought stress. And
promotion of g
s
by Gm relative to nonAM plants
occurred in the NaCl/drought stress treatment (28%
promotion) but not with exposure to drought alone
or to leaching of soil before drought.
The opposite of Hypothesis 2 occurred in four
instances: AM effects were more evident with
drought alone than with exposure to drought in
salinized soils. Leaf Chad declined to the same
extent in Gi and nonAM plants by the time the two
salinized soils had dried sufficiently to close
stomates. Yet in the unsalinized soil, stomates of
Gi plants reached the closure point at higher leaf C
than nonAM plants. Similarly, Gi and nonAM plants
reached stomatal closure at similar soil Cin each of
the salinized soils, but Gi plants reached stomatal
closure at much higher soil Cthan nonAM plants in
the unsalinized soil. Gi and nonAM plants reached
stomatal closure at the same time in salinized soils,
but Gi plants were able to maintain stomatal
opening 2 d longer than nonAM plants with drought
alone. Gm and nonAM plants reached stomatal
closure at similar soil Cin salinized soil, but Gm
plants reached stomatal closure in this instance at
much lower soil Cthan nonAM plants in the
unsalinized soil.
We were unable to test Hypothesis 2 in two
instances because leaves or soil had become so
dehydrated that we could not obtain reliable
ARTICLE IN PRESS
K. Cho et al.526

measurements: soil Cand leaf C
p
at the lethal
point for plants in unsalinized soil.
Our surmise that AM-induced salinity resistance
might help explain the observation that AM plants
are often more resilient to drought stress than their
nonAM counterparts does not appear to hold much
merit. In tests with different AM symbionts and
different ways of salinizing soils, the presence of
excess salt in soils widened the difference in
drought responses between AM and nonAM plants
only occasionally. In twice as many instances,
salinity stress tended to nullify an AM-induced
change in drought response.
Acknowledgments
This manuscript is based upon work supported by
the US Department of Agriculture under Award no.
00-35100-9238 and by the Tennessee Agricultural
Experiment Station.
References
Al-Karaki GN. Growth of mycorrhizal tomato and mineral
acquisition under salt stress. Mycorrhiza 2000;10:
51–4.
Al-Karaki GN, Hammad R, Rusan M. Response of two
tomato cultivars differing in salt tolerance to inocula-
tion with mycorrhizal fungi under salt stress. Mycor-
rhiza 2001;11:43–7.
Allen MF, Boosalis MG. Effects of two species of VA
mycorrhizal fungi on drought tolerance of winter
wheat. New Phytol 1983;93:67–76.
Auge´ RM. Water relations, drought and VA mycorrhizal
symbiosis. Mycorrhiza 2001;11:3–42.
Auge´ RM, Brown MS, Bethlenfalvay GJ, Stodola AJW.
Stomatal response of mycorrhizal cowpea and soybean
to short-term osmotic stress. New Phytol
1992;120:117–25.
Auge´ RM, Kubikova E, Moore JL. Foliar dehydration
tolerance of mycorrhizal cowpea, soybean and bush
bean. New Phytol 2001a;151:535–41.
Auge´ RM, Stodola AJW, Tims JE, Saxton AM. Moisture
retention properties of a mycorrhizal soil. Plant Soil
2001b;230:87–97.
Auge´ RM, Moore JL, Cho K, Stutz JC, Sylvia DM, Al-Agely
A, Saxton AM. Relating dehydration resistance of
mycorrhizal Phaseolus vulgaris to soil and root
colonization by hyphae. J Plant Physiol 2003;160:
1147–56.
Auge´ RM, Moore JL, Sylvia DM, Cho K. Mycorrhizal
promotion of host stomatal conductance in relation
to irradiance and temperature. Mycorrhiza
2004a;14:85–92.
Auge´ RM, Sylvia DM, Park SJ, Buttery BR, Saxton AM,
Moore JL, Cho K. Partitioning mycorrhizal influence on
water relations of Phaseolus vulgaris into soil and
plant components. Can J Bot 2004b;82:503–14.
Azco´n R, El-Atrash F. Influence of arbuscular mycorrhizae
and phosphorus fertilization on growth, nodulation
and N
2
(N-15) in Medicago sativa at four salinity levels.
Biol Fert Soils 1997;24:81–6.
Ben Khaled L, Gomez AM, Ouarraqi EM, Oihabi A.
Physiological and biochemical responses to salt stress
of mycorrhized and/or nodulated clover seedlings
(Trifolium alexandrinum L.). Agronomie 2003;23:
571–80.
Chapman DS, Auge´ RM. Drought resistance strategies of
four native ornamental perennials. J Am Soc Hort Sci
1994;119:299–306.
Chapman HD, Pratt PF. Methods of analysis for soils,
plants and waters. Riverside, CA: University of
California; 1961.
Davies FT, Potter JR, Linderman RG. Drought resistance
of mycorrhizal pepper plants independent of leaf
P-concentration –response in gas exchange and water
relations. Physiol Plant 1993;87:45–53.
Ebel RC, Welbaum GE, Gunatilaka M, Nelson T, Auge´ RM.
Arbuscular mycorrhizal symbiosis and nonhydraulic
signaling of soil drying in Vigna unguiculata (L.) Walp.
Mycorrhiza 1996;6:119–27.
Feng G, Zhang FS, Li XL, Tian CY, Tang C, Rengel Z.
Improved tolerance of maize plants to salt stress by
arbuscular mycorrhiza is related to higher accumula-
tion of soluble sugars in roots. Mycorrhiza 2002;12:
185–90.
Giri B, Mukerji KG. Mycorrhizal inoculant alleviates salt
stress in Sesbania aegyptiaca and Sesbania grandiflora
under field conditions: evidence for reduced sodium
and improved magnesium uptake. Mycorrhiza 2004;14:
307–12.
Hetrick BAD, Gerschefske K, Wilson GT. Effects of drought
stress on growth response in corn, sudan grass, and big
bluestem to Glomus etunicatum. New Phytol
1987;105:403–10.
Jindal V, Atwal A, Sekhon BS, Rattan S, Singh R. Effect of
vesicular-arbuscular mycorrhizae on metabolism of
moong plants under NaCl salinity. Plant Physiol
Biochem 1993;31:475–81.
Pfeiffer CM, Bloss HE. Growth and nutrition of
guayule (Parthenium argentatum) in a saline soil as
influenced by vesicular-arbuscular mycorrhiza and
phosphorus fertilization. New Phytol 1987;108:
315–21.
Porcel R, Barea JM, Ruiz-Lozano JM. Antioxidant activ-
ities in mycorrhizal soybean plants under
drought stress and their possible relationship to the
process of nodule senescence. New Phytol 2003;157:
135–43.
Rayment GE, Higginson FR. Australian laboratory hand-
book of soil and water chemical methods. Melbourne:
Inkata Press; 1992.
Rosendahl CN, Rosendahl S. Influence of vesicular-
arbuscular mycorrhizal fungi (Glomus spp.) on the
response of cucumber (Cucumis sativis L.) to salt
stress. Environ Exp Bot 1991;31:313–8.
ARTICLE IN PRESS
Mycorrhizal sorghum, salinity and drought stress 527

Ruiz-Lozano JM, Azco´n R. Symbiotic efficiency and
infectivity of an autochthonous arbuscular mycorrhi-
zal Glomus sp. from saline soils and Glomus deserti-
cola under salinity. Mycorrhiza 2000;10:137–43.
Ruiz-Lozano JM, Azco´n R, Gomez M. Alleviation of
salt stress by arbuscular-mycorrhizal Glomus species
in Lactuca sativa plants. Physiol Plant 1996;98:
767–72.
Simpson D, Daft MJ. Effects of Glomus clarum and water
stress on growth and nitrogen fixation in 2 genotypes
of groundnut. Agric Ecosyst Environ 1991;35:47–54.
Stirzaker RJ, Passioura JB. The water relations of the
root–soil interface. Plant Cell Environ 1996;19:201–8.
Subramanian KS, Charest C. Arbuscular mycorrhizae and
nitrogen assimilation in maize after drought and
recovery. Physiol Plant 1998;102:285–96.
ARTICLE IN PRESS
K. Cho et al.528