Annals of Botany 112: 429 –437, 2013
doi:10.1093/aob/mct069, available online at www.aob.oxfordjournals.org
PART OF A SPECIAL ISSUE ON MATCHING ROOTS TO THEIR ENVIRONMENT
Root cortical burden influences drought tolerance in maize
Raúl E. Jaramillo, Eric A. Nord, Joseph G. Chimungu, Kathleen M. Brown and Jonathan P. Lynch*
Department of Plant Science, The Pennsylvania State University, 102 Tyson Building, University Park, PA 16803, USA
* For correspondence. E-mail [email protected]
Received: 28 September 2012 Revision requested: 13 November 2012 Accepted: 7 February 2013 Published electronically: 25 April 2013
† Background and Aims Root cortical aerenchyma (RCA) increases water and nutrient acquisition by reducing the
metabolic costs of soil exploration. In this study the hypothesis was tested that living cortical area (LCA; transversal root cortical area minus aerenchyma area and intercellular air space) is a better predictor of root respiration,
soil exploration and, therefore, drought tolerance than RCA formation or root diameter.
† Methods RCA, LCA, root respiration, root length and biomass loss in response to drought were evaluated in
maize (Zea mays) recombinant inbred lines grown with adequate and suboptimal irrigation in soil mesocosms.
† Key Results Root respiration was highly correlated with LCA. LCA was a better predictor of root respiration
than either RCA or root diameter. RCA reduced respiration of large-diameter roots. Since RCA and LCA
varied in different parts of the root system, the effects of RCA and LCA on root length were complex.
Greater crown-root LCA was associated with reduced crown-root length relative to total root length. Reduced
LCA was associated with improved drought tolerance.
† Conclusions The results are consistent with the hypothesis that LCA is a driver of root metabolic costs and may
therefore have adaptive significance for water acquisition in drying soil.
Key words: Root, aerenchyma, respiration, drought, cortex, Zea mays.
IN T RO DU C T IO N
Drought is a primary limitation to global food production
(Boyer, 1982; Tuberosa et al., 2007). Moreover, crop drought
stress is projected to intensify in important agroecosystems
as a result of global climate change (Trenberth et al., 2007).
Root architecture is an important regulator of water acquisition
under drought. Plants with longer and deeper roots have
better access to water resources available at depth, and are therefore more prevalent among species found in dry environments
(Sharp and Davies, 1985; Merrill et al., 2002; Ho et al.,
2005; Moroke et al., 2005; Schenk and Jackson, 2005;
Morison et al., 2008). While plants generally allocate relatively
more resources to the root system in response to mineral
deficiencies and drought (Lynch, 2007a, b), the absolute
size of the root system may still be reduced. The physical location of roots is clearly important – roots are the pathways of
water and nutrient uptake, and traits that optimize the
co-location of roots with available resources will increase resource acquisition (Passioura, 1983; Ludlow and Muchow,
1990; Blum, 1996; Lynch, 1998, 2007a, b; Ho et al., 2004;
Benjamin and Nielsen, 2006; Tuberosa and Salvi, 2006;
Songsri et al., 2008).
Efficient utilization of metabolic resources for soil exploration
is a key aspect of plant adaptation to soil resource limitation
(Lambers et al., 2002; Lynch and Ho, 2005; Lynch, 2007a, b).
Metabolic resources allocated to the root have varying effectiveness depending on physiological, anatomical and architectural traits that affect their utilization (Lynch, 2007a, b). Both
construction and maintenance costs have an impact on the
standing biomass of root systems, the volume of soil explored,
and acquisition of soil resources (Lambers et al., 2002). This
has been demonstrated in the case of phosphorus (P) (Nielsen
et al., 2001; Lynch and Ho, 2005) and nitrogen (N) acquisition
(Robinson, 2001; King et al., 2003). For example, increased
specific root length (root length per unit mass) reduces
root costs and increases soil exploration, especially in
nutrient-limited environments where soil exploration is
favoured instead of root longevity (Ryser and Lambers, 1995;
reviewed by Ryser, 2006). A related mechanism to decrease
the metabolic cost of soil exploration and augment the allocation of resources to the growing points of the roots is root etiolation, the preferential increase in root elongation at the expense
of secondary development (De la Riva, 2010).
Root cortical aerenchyma (RCA), i.e. enlarged air spaces in
root cortical tissue (Evans, 2004; Gregory, 2006), is formed
constitutively and in response to a variety of stimuli including
hypoxia, excess temperature, deficiency of N, S and P, drought,
and mechanical impedance (Drew et al., 2000; Evans,
2004; Bouranis et al., 2006; Zhu et al., 2010). RCA disproportionately reduces root respiration in both P-deficient and
ethylene-induced aerenchymatous roots of maize (Zea
mays L.) at the scale of root segments and whole plants (Fan
et al., 2003), and is associated with improved growth in
low-P soil (Lynch, 2011). Studies using the functional –
structural plant model SimRoot showed that RCA formation
in common bean can increase P acquisition from low-P soil
and that RCA formation in maize can increase acquisition of
N, P and potassium by reducing the respiration and nutrient
content of root tissue (Postma and Lynch 2011a, b). Among
contrasting maize inbreds, high RCA formation reduced root
respiration, increased rooting depth, improved leaf water
status, increased plant biomass, and substantially improved
yield under drought (Zhu et al., 2010).
# The Author 2013. Published by Oxford University Press on behalf of the Annals of Botany Company. All rights reserved.
For Permissions, please email: [email protected]
430
Jaramillo et al. — Living cortical burden and drought tolerance
By converting living root cortical cells to air space, RCA
reduces the metabolic cost of soil exploration and therefore
can improve soil resource capture. This suggests that the metabolic demand of living cortical tissue is an important determinant of root growth, soil exploration and soil resource
capture. We propose that living cortical area (LCA, total cortical area minus RCA and intercellular air space) is a better
index of root economy than RCA or root diameter alone.
In the current study, we have quantified the effect of aerenchyma and root diameter on root respiration, root length and
root-type distribution in maize plants grown under wellwatered and water-stressed conditions. In addition to confirming a negative relationship between RCA and root respiration,
we also corroborated the benefit of thinner roots in reducing
the cost of root maintenance. We have demonstrated that
LCA is an important determinant of soil exploration and
water capture from drying soil.
M AT E R IA L S A ND M E T HO DS
Plant growth conditions
We conducted three experiments in temperature-controlled
greenhouses at University Park, PA, USA (40885’N,
77883′ W) in 2006, 2007 and 2011. The first two experiments
included six maize (Zea mays) genotypes, with contrasting formation of RCA, selected based on preliminary experiments
(Jaramillo-Velastegui, 2011) in which we screened 21 recombinant inbred lines (RILs) from the intermated B73 × Mo17
(IBM) population (Lee et al., 2002; Sharopova et al., 2002).
The third experiment included nine RILs from the IBM population. In the first experiment, we tested these six contrasting
RILs under moderate drought conditions, and measured root
respiration in nodal roots, using four replicates. In the
second experiment we tested four RILs under more severe
drought conditions and measured root respiration in primary,
seminal and nodal root samples. While five replicates were
grown in this experiment, root anatomical measurements
were only carried out for three replicates. In the third experiment, root anatomy and plant biomass data were collected
from four replicates.
In all studies, seeds were surface sterilized with 10 %
NaOCl for 1 min and rinsed with distilled water for 2 min.
Seeds were germinated in 0.5 mM CaSO4 for 48 h at 28 8C.
Germinated seeds were then transplanted to mesocosms consisting of 155-cm-tall PVC cylinders 15.7 cm in diameter
lined with sleeves made of 4-mil (0.116-mm) transparent
hi-density polyethylene film. Each mesocosm received two
germinated seeds, which were thinned to one plant per mesocosm after 2 d. The growth medium consisted of a mixture
(volume based) of 50 % medium size (0.5 – 0.3 mm) commercial grade sand (Quikrete Companies Inc., Atlanta, GA, USA),
35 % horticultural vermiculite (Whittemore Companies Inc.,
Lawrence, MA, USA), 5 % perlite (Whittemore Companies
Inc.) and 10 % topsoil. The topsoil was collected from a highfertility plot in the Russell E. Larson Agricultural Research
Center in Rock Springs, PA (Hagerstown silt loam; fine,
mixed, semiactive, mesic Typic Hapludalf, pH ≈ 6.7).
A uniform volume (28.5 L) of the soil mixture was used in
each mesocosm to ensure the same bulk density of the
media. Before transplanting, all mesocosms were saturated
with nutrient solution adjusted to pH 6.5 and consisting of
(in mM): KNO3 (3000), Ca(NO3)2 (1200), NH4NO3 (400),
MgSO4 (500), (NH4)2SO4 (250), KH2PO4 (100), KCl (50),
H3BO3 (25), MnSO4 (2), ZnSO4 (2), CuSO4 (0.5), (NH4)6
Mo7O24 (0.5) and Fe-Na-EDTA (50). The mesocosms were
allowed to drain to assure field capacity at planting. In the
first experiment the initial saturation was 8 L of nutrient solution, and in the second and third experiment this was reduced
to 4.8 L. In the first experiment, the control mesocosms
received 100 mL of deionized water daily, and the drought
mesocosms received 100 mL three times per week for the
first 2 weeks, and no irrigation in the second 2 weeks. To
increase the drought stress in the second experiment, the
drought mesocosms received 50 mL of deionized water every
other day for the first week and then no additional irrigation,
while the control mesocosms received 100 mL of deionized
water every other day. In the third experiment the control
mesocosms received 200 mL of deionized water every day,
while the drought mesocosms received 100 mL daily for the
first 4 d, and then received no further irrigation.
The experiments were carried out with an average day-time
natural light of 450 mmol photons m22 s21 PAR, with
maximum of 1200 mmol photons m22 s21 PAR; additional
light was provided with 400-W metal-halide bulbs (Energy
Technics, York, PA, USA) for 14 h per day. Total PAR averaged 30.24 mol photons m22 d21. Average daytime temperature in the greenhouse was approximately 28 8C.
Harvesting and measurement of root respiration
and net carbon exchange
Whole root respiration and net shoot carbon assimilation
were measured 1 d before harvesting, 4 weeks after planting.
We used a system of chambers connected to a Li-6200 infrared
gas analyser (Li-Cor Environmental Inc.). In short, a 37-L
(45 × 30 × 28 cm) transparent acrylic chamber (with a split
base designed to seal around the stem of a plant) was placed
around a single plant. The base of the chamber was carefully
sealed around the stem of the plant with modelling clay to separate the headspace over the growth media from the enclosed
airspace around the shoot. The chamber was hermetically
sealed and connected to the Li-6200 with polyethylene
tubing, and the air in the chamber was continually mixed
with a small fan. The base of the chamber formed a seal
against the top of the mesocosm, and was also fitted with polyethylene tubing connected to the Li-6200 to measure the
respiration of the whole root system and the media. We
employed the whole root system plus media respiration as a
proxy for total root respiration, assuming that the amount of
natural soil and the respiration of microbes is the same in all
cylinders (Bouma et al., 1997a, b). Carbon dioxide exchange
of roots and shoots was measured for 1 min and 2 min, respectively, for each plant. The total decrease or increase of CO2
concentration, barometric pressure and temperature and
volume of the chambers used, were employed to estimate
shoot photosynthesis (mmol CO2 g21 leaf dry matter s21) or
CO2 produced in the media (mmol CO2 s21). The whole root
system respiration values were divided by the total root
length obtained by WinRhizo scanning (described below) to
Jaramillo et al. — Living cortical burden and drought tolerance
obtain the specific root respiration rate per unit of root length
(mmol CO2 m21 root s21). Instantaneous leaf-level photosynthesis and transpiration were measured using a Li-6400 portable photosynthesis system (Li-Cor Biosciences, Lincoln,
NE, USA).
Four weeks after planting, plants were harvested by cutting
off the shoot and removing the plastic sleeve from the supporting PVC cylinder. The plastic sleeve was then cut open and the
roots were recovered by carefully rinsing the media away with
water. Root segments, 20 cm long, were collected at 20– 40 cm
from the root tip for respiration measurements. In the first experiment these were taken only from the two longest nodal
roots. In the second experiment samples were also taken
from one seminal root and the primary root. The lateral roots
in these samples were removed with a Teflon-coated razor
blade and the samples were patted dry and placed in a
40-mL custom chamber connected to the Li-6200 infrared
gas analyser. The container with the sample was kept at a
stable temperature at 18 8C for 2 min while the respiration
was measured. The remainder of the root system was stored
in 25 % ethanol for further processing. The shoots were oven
dried at 60 8C for 2 – 3 d for dry-matter determination.
Root anatomical observations and root length
Root anatomy was characterized based on free-hand crosssections cut from root segments approx. 5 cm long, taken
from the root sample used for respiration measurement (a
20-cm sample taken 20– 40 cm from the tip of the longest
nodal root present). The root segments were placed in a drop
of water over dental wax and cut with Teflon-coated razor
blades under a dissecting microscope [ protocol suggested by
Rosemary White (CSIRO, Australia; http://roots.psu.edu/
?q=en/node/151)]. The cross-sections were imaged with a
Hamamatsu C2400 CCD camera attached to a Nikon
Diaphot inverted microscope using ImageMaster (Photon
Technology International, Birmingham, NJ, USA). Images
were analysed using Image-J (http://rsb.info.nih.gov/ij/) and
using RootScan, an image analysis tool developed for analysing root anatomy (Burton et al., 2012). The area (mm2) of the
total root section, of the stele and of RCA was measured in 6 –
Living cortical area (mm2)
1·5
A
431
12 different images per sample. Cortex area was obtained by
subtracting the stele area from the total root section area and
LCA was obtained by subtracting RCA from the cortex area.
The percentage of cortical area occupied by aerenchyma was
calculated by dividing the RCA area by the cortex area.
The rest of the root system was separated into primary,
seminal and nodal roots. The length for each root type was
quantified with a WinRhizo scanning system (Regent
Instruments Inc., Quebec, Canada). Fine lateral root length
was separated from main axis root length within each root
class based on root diameter.
Statistical analysis
For comparisons between high and low RCA groups (or individual RILs), irrigation levels and the interaction between
these main effects a two-way ANOVA was used. A protected
least significant difference post hoc (a ¼ 0.05) test was used
for multiple comparisons. We carried out correlations and
linear regressions between shoot and root traits with RCA,
LCA and root section area. Data were analysed using R
version 2.11 (R Development Core Team, 2010), with
ANOVAs fitted using the function ‘Anova’ from the package
‘car’ (Fox and Weisberg, 2010).
R E S ULT S
LCA is influenced by root diameter and RCA
LCA was strongly correlated with root diameter in both 2006
and 2007 (Fig. 1A), though the relationship differed between
experiments (Table 1, P , 0.001). The relationship between
RCA and LCA was quite weak (R 2 , 0.14; Fig. 1B), but significant (P ¼ 0.0065 in 2006 and P ¼ 0.0337 in 2007). LCA
was best predicted by including both root diameter and RCA
(R 2 ¼ 0.989 for both years; Table 1).
Response of RCA and LCA to drought
Drought treatment slightly decreased RCA (from 4.5 % to
3.2 %), but did not alter LCA (Table 2). In 2007 there were
B
Control 2006
Drought 2007
Control 2007
Drought 2007
1·0
0·5
0
0·4
0·6
0·8
1·0
1·2
Root diameter (mm)
1·4
1·6 0
5
10
15
Root cortical aerenchyma (%)
20
F I G . 1. Relationship of living cortical area (LCA) with (A) root diameter (P , 0.0001, R 2 ¼ 0.982 in 2006; P , 0.0001, R 2 ¼ 0.972 in 2007), and with
(B) root cortical aerenchyma (P ¼ 0.006, R 2 ¼ 0.137 in 2006; not shown in 2007–P ¼ 0.10, R 2 ¼ 0.03).
432
Jaramillo et al. — Living cortical burden and drought tolerance
differences in RCA between the different root classes for RIL
77; percentage RCA was greater in the primary roots than in
crown or seminal roots (data not shown), but these differences
did not extend to the other genotypes, and the response of
RCA to drought treatment did not differ for root class.
combinations of genotype and root class that could be classed
as ‘high LCA’ (crown roots from RILs 304, 331, 364 and
primary roots from RIL 364) and ‘low LCA’ (seminal and
primary roots from RILs 304 and 77). Root-segment respiration decreased from 0.519 to 0.378 mmol CO2 s cm21 for
the ‘low LCA’ group (P ¼ 0.021).
LCA, RCA and root respiration
Root-segment respiration was not well correlated with RCA
in 2007 (Table 3). Root diameter was a stronger predictor than
RCA of root-segment respiration (Fig. 2A and Table 3). LCA
was also a predictor of root-segment respiration (Table 3), with
a slightly greater coefficient of determination than RCA (0.34
vs. 0.27). However, RCA was associated with substantial
reductions in respiration of large-diameter roots, but not in
small-diameter roots (Fig. 3).
Root-system respiration (estimated by root and soil CO2 production) was not correlated with RCA in 2007 (Table 3).
Increasing LCA or root diameter were both associated with
greater root-system respiration, with root diameter having a
slightly greater coefficient of determination (0.45 vs. 0.40;
Table 3).
Although the relationship between LCA and the respiration
of root segments was quite strong (Fig. 2), LCA was quite variable in the RILs we used (Fig. 4) – variable both among root
classes within a genotype (e.g. RIL 304 and RIL 77) and
within a root class for some genotypes (e.g. RIL 331).
Comparing across genotypes, crown-root LCA tended to be
greater than seminal-root LCA, and primary-root LCA was
most variable (Fig. 4). In the 2007 experiment, we did identify
TA B L E 1. Summary of a multiple regression model of LCA as
predicted by root diameter and RCA in six maize RILs in 2006
and and four maize RILs in 2007
Effect
Root diameter
RCA (%)
Year
Root diameter × year
R2
d.f.
F-value
(1,100)
(1,100)
(1,100)
(1,100)
0.989
5740***
64.9***
3.78†
766***
The model was LCA ¼ root diameter + RCA + year + root diameter ×
year.
†
P , 0.1; ***P , 0.001. Degrees of freedom are shown as (numerator,
denominator).
RCA, LCA and root growth
Root length was not affected by RCA when high- and
low-RCA RILs were compared (Table 4) or when percentage
RCA was considered (Table 5). However, both RCA and
LCA were associated with relative length of different segments
of the root system. Genotypes with greater crown-root RCA
had greater primary root-length fraction (Table 4). Greater
crown-root LCA was associated with a decrease in crown
root-length fraction (Table 6), and with a marginal increase
in seminal root length fraction (P ¼ 0.072; Table 4).
Drought treatment reduced the total root length in both years
(Fig. 5 and Table 4). When the 2007 experiment was considered alone, LCA was positively associated with total root
length (P , 0.001; data not shown).
Relationships of RCA and LCA with above-ground biomass
and drought tolerance
There was no difference in the above-ground biomass of
RILs classified as high RCA and low RCA (Table 4). There
TA B L E 3. Summary of linear models of root respiration as
predicted by root diameter, RCA and LCA in four maize RILs
under moderate drought and control conditions in 2007
RCA (%)
Root diameter (mm)
Root-segment respiration (mmol CO2 s21)
Anatomy
1.62 (1,93)
34.4 (1,93)***
Irrigation
3.17 (1,93)
1.59 (1,93)
0.0231
0.269
R2
Root system respiration (ppm CO2 s21)
Anatomy
1.96 (1,16)
7.28 (1,16)*
Irrigation
8.28 (1,16)*
7.18 (1,16)*
0.292
0.450
R2
LCA (mm2)
46.4 (1,93)***
3.39 (1,93)
0.337
5.01 (1,16)*
8.08 (1,16)*
0.395
Data shown are F-values with numerator and denominator degrees of
freedom in parenthesis.
*P , 0.05; ***P , 0.001.
TA B L E 2. Summary of linear models of RCA (% cortex), LCA and root diameter in six maize RILs in 2006 and and four maize
RILs in 2007
LCA (mm2)
RCA (% cortex)
Effect
Genotype
Irrigation
Year
Genotype × year
R2
†
Root diameter (mm)
d.f.
F-value
d.f.
F-value
d.f.
F-value
(5,59)
(1,59)
(1,59)
–
0.231
3.73**
2.18
8.0**
–
(5,57)
–
(1,57)
(3,57)
0.613
1.29
–
80.44***
3.06*
(5,58)
(1,58)
(1,58)
(3,58)
0.029
0.59
1.06
0.39
2.54†
P , 0.1; *P , 0.05; **P , 0.01; ***P , 0.001. Degrees of freedom are shown as (numerator, denominator).
Root segment respiration (nmol CO2 s–1)
Jaramillo et al. — Living cortical burden and drought tolerance
1·2
A
B
2006
2007
1·0
433
0·8
0·6
0·4
0·2
0
0
0·5
1·0
1·5
Root diameter (mm)
1·5
0·5
1·0
Living cortical area (mm2)
0
Living cortical area (mm2)
0·06
0·05
0·04
0·03
Crown
Seminal
Primary
1·5
1·0
304
r (m
1·5
1·4
1·3
1·2
1·1
1·0
m)
0·5
0·02
5
a (%)
erenchym
iam
10
Cortical a
0
ot d
15
F I G . 3. Response surface of the respiration of root segments from six IBM
RILs with root cortical aerenchyma (RCA) formation and root diameter. The
surface was generated using a loess smoother (using the R function loess
with degree 2 and span 0.75r). The surface reflects 44 points, the two points
with highest RCA (18.6 %) and largest root diameter (1.7 mm) were excluded
as the density of data at those values was very low.
was, however, a decrease in shoot biomass associated with increasing percentage RCA, and an increase in shoot biomass
associated with increased LCA (Table 5). Both of these
trends were significantly stronger in 2007 (Table 5).
To assess drought tolerance, shoot biomass in the drought
treatment was normalized by shoot biomass in the control
treatment. There was no relationship between percentage
RCA and relative shoot biomass, but genotypes with greater
LCA were more sensitive to drought, as shown by relative
shoot biomass (Fig. 6).
Transpiration was not systematically affected by RCA,
percentage RCA, LCA or drought treatment (Fig. 7). There
were significant differences in transpiration associated with
genotype, and increasing percentage RCA decreased transpiration in water-stressed, but not in well-watered, plants. Leaf
331
364
77
RIL
ete
0·01
Ro
Root segment respiration
(nmol CO2 s
–1
cm–1)
F I G . 2. Root-segment respiration from root segments is correlated with root diameter (A) and LCA (B).
F I G . 4. Living cortical area (LCA) in three root classes from four maize RILs
grown with sufficient (wide bars) and suboptimal (narrow bars) irrigation in
mesocosms in 2007. LCA in crown roots in 2006 was generally similar to
that in 2007. Upper and lower edges of the boxes show the 1st and 3rd quartiles. Vertical dashed lines separate the four RILs.
TA B L E 4. Summary of linear models (F-value and degrees of
freedom) of the responses of root length and above-ground
biomass to root cortical aerenchyma (RCA) in four maize RILs
under moderate drought and control conditions in 2007
Effect
RCA
Irrigation
Year
R2
Total root
length
Primary root
fraction
Above-ground
biomass
2.56 (1,42)
12.8 (1,42)***
129 (1,42)***
0.846
5.03 (1,42)*
0.0729 (1,42)
40.1 (1,42)***
0.437
0.016 (1,58)
20.5 (1,58)***
4.70 (1,58)*
0.278
*P , 0.05; ***P , 0.001. Non-significantly different group means
indicated by a dash. Degrees of freedom are shown as (numerator,
denominator).
area was measured only in the second experiment, and
was reduced in drought-treated plants (P , 0.01, data not
shown).
TA B L E 6. Summary of ANCOVA models (F-value and degrees of
freedom) of root length fraction as influenced by LCA and
irrigation in in six maize RILs in 2006 and and four maize RILs
in 2007
Crown
Seminal
Primary
6.25 (1,60)*
2.62 (1,60)
0.761 (1,60)
0.110
3.37 (1,60)†
1.68 (1,60)
10.0 (1,60)**
0.436
1.71 (1,60)
0.554 (1,60)
16.5 (1,60)
0.250
†
P , 0.1, *P , 0.05, **P , 0.01. Degrees of freedom shown as
(numerator, denominator).
Total root length (cm)
8000
Crown laterals
Seminal laterals
Primary laterals
Main axis
6000
4000
2000
0
C
C
D
304
D
C
D
331
364
RIL and irrigation
C
D
77
F I G . 5. Main axis (hatched) and lateral root length for the three classes of
roots in four maize RILs grown with sufficient (C) and deficient (D) irrigation
in 2007.
Relative biomass (drought : control)
P , 0.1, *P , 0.05, **P , 0.01, ***P , 0.001. Non-significant interactions are indicated by a dsh. Degrees of freedom shown as (numerator, denominator).
LCA
Irrigation
Year
R2
†
6.70 (1,61)*
29.0 (1,61)***
19.3 (1,61)***
32.0 (1,61)***
0.558
6.44 (1,61)*
26.6 (1,61)***
5.43 (1,61)*
4.06 (1,61)*
0.384
0.0224 (1/61)
10.4 (1/61)**
181 (1/61)***
2.95 (1/61)†
0.748
0.0984 (1,60)
11.0 (1,60)**
74.2 (1,60)***
–
0.748
1.52 (1,60)
12.2 (1,60)***
177 (1,60)***
–
0.754
Covariate
Irrigation
Year
Covariate × year
R2
LCA
Effect
RCA
LCA
Root diameter
RCA
Above-ground biomass
Effect
Total root length
TA B L E 5. Summary of ANCOVA models (F-value and degrees of freedom) of the responses of root length and shoot biomass to RCA, LCA and root diameter
in six maize RILs in 2006 and and four maize RILs in 2007
9.71 (1/63)**
23.4 (1/63)***
12.6 (1/63)***
23.8 (1/63)***
0.516
Jaramillo et al. — Living cortical burden and drought tolerance
Root diameter
434
2007
2011
1·2
1·0
0·8
0·6
0·4
0·3
0·4
0·5
0·6
0·7
Crown root LCA (mm2)
0·8
0·9
F I G . 6. Relative shoot biomass (ratio of drought to control) as a function of
living cortical area of crown roots of maize grown in the greenhouse. Four
RILs were grown in 2007 (P ¼ 0.058, R 2 ¼ 0.34) and nine RILs in 2011
(P ¼ 0.008, R 2 ¼ 0.18). Relative biomass was calculated for each RIL in
each replicate.
D IS C US S IO N
Previous studies have associated reduced root metabolic costs
with adaptation to low P availability. The production of RCA
(Fan et al., 2003), the relative abundance of metabolically lessdemanding adventitious roots in bean (Miller et al., 2003) and
a decrease in root respiration (Nielsen et al., 2001; Walk et al.,
Jaramillo et al. — Living cortical burden and drought tolerance
10
A
24
20
6
16
12
4
8
Percent RCA
Shoot dry mass (g)
8
2
4
0
2·0
Control
Drought
Transpiration (mmol m–2 s–1)
B
1·5
1·0
0·5
0
331
304
364
77
RIL
F I G . 7. (A) Shoot biomass and percentage root cortical aerenchyma, and (B)
transpiration in maize RILs grown in the greenhouse with adequate
(‘control’) or deficient (‘drought’) irrigation. Superimposed rectangles (A)
demonstrate percentage RCA + s.e., which was determined from crosssections of crown roots.
2006) have been related to longer roots and reduced P stress.
RCA formation has also been associated with deeper rooting
under drought and increased drought tolerance in maize (Zhu
et al., 2010). Simulation studies with maize and bean found
that RCA increased soil exploration and nutrient capture
(Postma and Lynch, 2011a, b). Our study confirms the benefits
of reduced root metabolic costs for water acquisition and reduction of drought stress, and identifies LCA as an important
aspect of root metabolic cost. We show that metabolically efficient roots reduce the effects of drought stress as they do for
P deficiency, but in this case by permitting access to deep soil
water.
In previous work, we have focused on the relationship
between RCA and root metabolic costs, reasoning that the
loss of cortical cells during RCA formation reduces maintenance respiration, freeing resources for root extension (Fan
et al., 2003; Zhu et al., 2010; Postma and Lynch, 2011a, b).
Here we focus on a major component of the cost of the
roots, LCA, representing the remaining living cortical cells.
435
LCA should be a more direct predictor of root respiratory
costs than RCA since it takes into account the differing cortical
areas among root classes, genotypes and treatments. LCA is
the cross-sectional area of living cortical cells. LCA should
be proportional to root diameter (Fig. 1A) with error resulting
from variation in stele area, non-circularity of roots and measurement error, and LCA should be inversely proportional to
RCA (Fig. 1B). Since stele area was a relatively low fraction
of the root cross-section area (17 %) and was not highly variable (95 % of values fell between 13 and 22 % of root crosssectional area), LCA should be very well predicted by root
diameter and RCA, as was the case in this study (Table 1).
The relationship between root metabolic costs and LCA
should be strongest in older roots, where RCA formation is
expected to be greatest. Conversely, in younger roots, metabolic costs should be more strongly related to root diameter. Our
results generally conform to this pattern. Root-segment respiration (of older, more basal root segments) was marginally
more correlated with LCA (Fig. 2 and Table 3), while whole
root-system respiration, which integrates roots with and
without RCA, was marginally more correlated with root diameter (Table 3). We do note that none of these models are highly
predictive, and differences between them are small. However,
their conformation to expectation is noteworthy. Variability in
root-segment respiration likely reflects both differences in stele
area and metabolic activity as well as measurement error.
Since variability in the stele area was not associated with irrigation treatment (data not shown), there is no reason to suspect
that these sources of error should be biased.
Zhu et al. (2010), working with some of the same genotypes
as in our studies, but focusing only on RCA formation,
observed in greenhouse and field environments that RCA formation improves drought tolerance. In our observations, genotypes with high RCA formation can have lower LCA than
would be expected based on root diameter (i.e. points that
fall below the regression lines in Fig. 1A correspond to
points that are toward the right edge of Fig. 1B). The RILs
used in these experiments did not show significant differences
in RCA area or LCA associated with water availability, but
there was a small decline in percentage RCA in water-stressed
plants (Fig. 7 and Table 1). If RCA formation increased under
drought, as reported by Zhu et al. (2010), then plants would be
able to reduce their cortical burden under stress and possibly
support increased root length. We note that Zhu’s results are
from field-grown plants harvested at 55– 70 d after planting,
while our plants were harvested at 28 d and experienced less
severe drought stress.
One possible mechanism by which lower LCA could increase drought tolerance is by permitting enhanced root
growth, as carbon and nutrient savings from reduced root
metabolic demand could be allocated into enhanced shoot
and root growth, and ultimately increase root depth. In these
experiments, total root length was not altered by RCA, percentage RCA or LCA (Table 3). There were, however, shifts
in root length among root classes associated with RCA and
LCA. Increased crown-root LCA was associated with
reduced relative crown-root length (Table 3). Similarly, genotypes with low crown-root RCA had reduced relative fine-root
length and relative primary-root length compared with high
RCA genotypes (Table 3), which agrees with previous
436
Jaramillo et al. — Living cortical burden and drought tolerance
reports that increasing RCA can enhance root growth (Fan
et al., 2003; Zhu et al. 2010). While we did not observe
increased root length associated with lower LCA, we note
that such a response would rely on growth feedback and so
may be difficult to detect as early as 28 d after planting. For
example, Postma and Lynch (2011b) reported a doubling of
the effect of RCA on total dry weight between 28 d and 40 d.
We did not observe a regular, predictable production of
RCA among the selected IBM RILs. For instance, RCA in
RIL 364 varied from 0 % to 15 % in these experiments
(Fig. 7). In the work of Zhu et al. (2010), RIL 364 produced
the greatest root length in deeper soil layers under drought in
the field, and also showed a substantial increase in RCA
under drought. RCA plays a role in the reduction of metabolic
costs with thick roots, e.g. IBM RIL 77 had thick crown roots
and relatively high and constant RCA in all tests; this material
might benefit from increased RCA. Large root diameter and
therefore high LCA may be advantageous in drying soil,
since drying soils often become harder and the ability to
penetrate hard soil is related to root diameter (Bengough
et al., 2011).
In this study, drought did not alter transpiration on a
leaf-area basis, but did reduce both above-ground biomass
and leaf area (Fig. 7). Considering both experiments with
these IBM RILs, RILs 364 and 77 have several traits that
reduce the metabolic cost of water acquisition, with moderate
to high production of RCA, and without major reductions in
RCA in drought conditions (Fig. 7). RIL 77 has relatively
low LCA for all root classes (Fig. 4). RILs 364 and 77 show
relatively minor reductions in above-ground biomass and leaf
area (Fig. 7), and only minor reductions in root length in response to drought (Fig. 5). Seminal roots have previously
been related to the drought response of maize (Sharp and
Davies, 1985) and other cereals have shown better adaptation
to drought when longer seminal roots are present. Seminal
root length has been associated with timing of water use
rather than metabolic efficiency (Richards and Passioura,
1981; Passioura, 1983; Richards, 2006).
In this study we have shown that while RCA formation can
reduce root respiration of large diameter roots, LCA is a stronger predictor of root respiration than either RCA or root diameter. We have also shown that plants with lower LCA suffered
less growth reduction by drought treatment. Since LCA is total
cross-sectional area less stele area less RCA less intracellular
air space, root diameter, stele diameter and RCA formation
all should affect LCA. The effect of these and other anatomical
traits on root respiration costs and plant tolerance to drought
and other edaphic stresses is worthy of further exploration.
AC KN OW LED GEMEN T S
This research was supported by the US National Science
Foundation/Basic Research to Enhance Agricultural
Development (grant no. 4184-UM-NSF-5380).
L I T E R AT U R E CI T E D
Bengough AG, McKenzie BM, Hallett PD, Valentine TA. 2011. Root elongation, water stress, and mechanical impedance: a review of limiting
stresses and beneficial root tip traits. Journal of Experimental Botany
62: 59– 68.
Benjamin JG, Nielsen DC. 2006. Water deficit effects on root distribution of
soybean, field pea and chickpea. Field Crops Research 97: 248– 253.
Blum A. 1996. Crop responses to drought and the interpretation of adaptation.
Plant Growth Regulation 20: 135– 148.
Bouma TJ, Nielsen KL, Eissenstat DM, Lynch JP. 1997a. Estimating respiration of roots in soil: interactions with soil CO2, soil temperature and soil
water content. Plant and Soil 195: 221– 232.
Bouma TJ, Nielsen KL, Eissenstat DM, Lynch JP. 1997b. Soil CO2 concentration does not affect growth or root respiration in bean or citrus. Plant,
Cell & Environment 20: 1495– 1505.
Bouranis DL, Chorianopoulou SN, Kollias C, et al. 2006. Dynamics of aerenchyma distribution in the cortex of sulfate-deprived adventitious roots
of maize. Annals of Botany 97: 695–704.
Boyer JS. 1982. Plant productivity and environment. Science 218: 443– 448.
Burton AL, Williams MS, Lynch JP, Brown KM. 2012. RootScan: software
for high-throughput analysis of root anatomical traits. Plant and Soil 357:
189– 203.
De la Riva LM. 2010. Root etiolation as a strategy for phosphorus acquisition
in common bean. MS Thesis, The Pennsylvania State University, USA.
Drew MC, He C, Morgan PW. 2000. Programmed cell death and aerenchyma
formation in roots. Trends in Plant Science 5: 123– 127.
Evans DE. 2004. Aerenchyma formation. New Phytologist 161: 35– 49.
Fan MS, Zhu JM, Richards C, Brown KM, Lynch JP. 2003. Physiological
roles for aerenchyma in phosphorus-stressed roots. Functional Plant
Biology 30: 493– 506.
Fox J, Weisberg S. 2010. car: Companion to Applied Regression. R package
version 2.0-2. http://www.inside-r.org/packages/car.
Gregory PJ. 2006. Plant roots: growth, activity and interactions with soils.
Oxford: Blackwell Publishing.
Ho MD, McCannon BC, Lynch JP. 2004. Optimization modeling of plant
root architecture for water and phosphorus acquisition. Journal of
Theoretical Biology 226: 331– 340.
Ho MD, Rosas JC, Brown KM, Lynch JP. 2005. Root architectural
tradeoffs for water and phosphorus acquisition. Functional Plant
Biology 32: 737– 748.
Jaramillo-Velastegui R. 2011. The edaphic control of plant response to
climate change: extent, interactions and mechanisms of plant adaptation.
PhD Thesis, The Pennsylvania State University, USA.
King J, Gay A, Sylvester-Bradley R, et al. 2003. Modelling cereal root
systems for water and nitrogen capture: towards an economic optimum.
Annals of Botany 91: 383–390.
Lambers H, Atkin O, Millenaar FF. 2002. Respiratory patterns in roots in
relation to their functioning. In: Waisel Y, Eshel A, Kafkaki K. eds.
Plant roots: the hidden half. New York, NY: Marcel Dekker, 521– 552.
Lee M, Sharopova N, Beavis WD, et al. 2002. Expanding the genetic map of
maize with the intermated B73 × Mo17 (IBM) population. Plant
Molecular Biology 48: 453–461.
Ludlow MM, Muchow RC. 1990. A critical-evaluation of traits for improving
crop yields in water-limited environments. Advances in Agronomy
43: 107 –153.
Lynch J. 1998. The role of nutrient efficient crops in modern agriculture.
Journal of Crop Production 1: 241–264.
Lynch JP. 2007a. Rhizoeconomics: the roots of shoot growth limitations.
Hortscience 42: 1107– 1109.
Lynch JP. 2007b. Roots of the second green revolution. Australian Journal of
Botany 55: 493–512.
Lynch JP. 2011. Root phenes for enhanced soil exploration and phosphorus
acquisition: tools for future crops. Plant Physiology 156: 1041–1049.
Lynch JP, Ho MD. 2005. Rhizoeconomics: carbon costs of phosphorus acquisition. Plant and Soil 269: 45– 56.
Merrill SD, Tanaka DL, Hanson JD. 2002. Root length growth of eight crop
species in haplustoll soils. Soil Science Society of America Journal
66: 913 –923.
Miller CR, Ochoa I, Nielsen KL, Beck D, Lynch JP. 2003. Genetic variation
for adventitious rooting in response to low phosphorus availability: potential utility for phosphorus acquisition from stratified soils. Functional
Plant Biology 30: 973– 985.
Morison JIL, Baker NR, Mullineaux PM, Davies WJ. 2008. Improving
water use in crop production. Philosophical Transactions of The Royal
Society B – Biological Sciences 363: 639– 658.
Jaramillo et al. — Living cortical burden and drought tolerance
Moroke TS, Schwartz RC, Brown KW, Juo ASR. 2005. Soil water depletion
and root distribution of three dryland crops. Soil Science Society of
America Journal 69: 197– 205.
Nielsen KL, Eshel A, Lynch JP. 2001. The effect of phosphorus availability
on the carbon economy of contrasting common bean (Phaseolus vulgaris
L.) genotypes. Journal of Experimental Botany 52: 329– 339.
Passioura JB. 1983. Roots and drought resistance. Agricultural Water
Management 7: 265–280.
Postma JA, Lynch JP. 2011a. Root cortical aerenchyma enhances the growth
of maize on soils with suboptimal availability of nitrogen, phosphorus,
and potassium. Plant Physiology 156: 1190– 1201.
Postma JA, Lynch JP. 2011b. Theoretical evidence for the functional benefit
of root cortical aerenchyma in soils with low phosphorus availability.
Annals of Botany 107: 829– 841.
R Development Core Team. 2010. R: a language and environment for statistical computing. R Foundation for Statistical Computing. http://
r-project.org.
Richards RA. 2006. Physiological traits used in the breeding of new cultivars
for water-scarce environments. 80: 197–211.
Richards RA, Passioura JB. 1981. Seminal root morphology and water-use
of wheat. 1. Environmental effects. Crop Science 21: 249– 252.
Robinson D. 2001. Root proliferation, nitrate inflow and their carbon costs
during nitrogen capture by competing plants in patchy soil. Plant and
Soil 232: 41– 50.
Ryser P. 2006. The mysterious root length. Plant and Soil 286: 1 –6.
Ryser P, Lambers H. 1995. Root and leaf attributes accounting for the performance of fast-growing and slow-growing grasses at different nutrient
supply. Plant and Soil 170: 251– 265.
437
Schenk HJ, Jackson RB. 2005. Mapping the global distribution of deep roots
in relation to climate and soil characteristics. Geoderma 126: 129–140.
Sharopova N, McMullen MD, Schultz L, et al. 2002. Development and
mapping of SSR markers for maize. Plant Molecular Biology 48:
463–481.
Sharp RE, Davies WJ. 1985. Root-growth and water-uptake by maize plants
in drying soil. Journal of Experimental Botany 36: 1441–1456.
Songsri P, Jogloy S, Vorasoot N, Akkasaeng C, Patanothai A, Holbrook
CC. 2008. Root distribution of drought-resistant peanut genotypes in
response to drought. Journal of Agronomy and Crop Science 194:
92–103.
Trenberth KE, Jones PD, Ambenje P, et al. 2007. Observations: surface and atmospheric climate change. In: Solomon S, Qin D, Manning M, et al eds.
Climate change 2007: the physical science basis. Contribution of Working
Group I to the Fourth Assessment Report of the Intergovernmental Panel
on Climate Change. Cambridge: Cambridge University Press.
Tuberosa R, Salvi S. 2006. Genomics-based approaches to improve drought
tolerance of crops. Trends in Plant Science 11: 405–412.
Tuberosa R, Salvi S, Giuliani S, et al. 2007. Genome-wide approaches to investigate and improve maize response to drought. Crop Science 47:
S120–S141.
Walk T, Jaramillo RE, Lynch J. 2006. Architectural tradeoffs between adventitious and basal roots for phosphorus acquisition. Plant and Soil
279: 347 –366.
Zhu JM, Brown KM, Lynch JP. 2010. Root cortical aerenchyma improves
the drought tolerance of maize (Zea mays L.). Plant, Cell &
Environment 33: 740 –749.