Journal of Plant Physiology 169 (2012) 1150–1157

Contents lists available at SciVerse ScienceDirect

Journal of Plant Physiology

journal homepage: www.elsevier.de/jplph

Flooding affects uptake and distribution of carbon and nitrogen in citrus seedlings
Belén Martínez-Alcántara 1 , Sara Jover 1 , Ana Quiñones, María Ángeles Forner-Giner,
Juan Rodríguez-Gamir, Francisco Legaz, Eduardo Primo-Millo, Domingo J. Iglesias ∗
Centro de Citricultura y Producción Vegetal, Instituto Valenciano de Investigaciones Agrarias, Ctra. Moncada-Náquera, Km. 4,5, 46113 Moncada, Valencia, Spain

a r t i c l e i n f o a b s t r a c t

Article history: Soil flooding has been widely reported to affect large areas of the world. In this work, we investigated
Received 18 October 2011 the effect of waterlogging on citrus carbon and nitrogen pools and partitioning. Influence on their uptake
Received in revised form 20 March 2012 and translocation was also studied through 15 N and 13 C labeling to provide insight into the physiological
Accepted 29 March 2012
mechanisms underlying the responses. The data indicated that flooding severely reduced photosynthetic
activity and affected growth and biomass partitioning. Total nitrogen content and concentration in the
Keywords:
plant also progressively decreased throughout the course of the experiment. After 36 days of treatment,
Carbon
nitrogen content of flooded plants had decreased more than 2.3-fold compared to control seedlings, and
Citrus
Flooding
reductions in nitrogen concentration ranged from 21 to 55% (in roots and leaves, respectively). Specific
Isotopic labeling absorption rate and transport were also affected, leading to important changes in the distribution of
Nitrogen this element inside the plant. Additionally, experiments involving labeled nitrogen revealed that 15 N
uptake rate and accumulation were drastically decreased at the end of the experiment (93% and 54%,
respectively).
13
CO2 assimilation into the plant was strongly reduced by flooding, with ␦13 C reductions ranging from
22 to 37% in leaves and roots, respectively. After 36 days, the relative distribution of absorbed 13 C was also
altered. Thus, 13 C recovery in flooded leaves increased compared to controls, whereas roots exhibited the
opposite pattern. Interestingly, when carbohydrate partitioning was examined, the data revealed that
sucrose concentration was augmented significantly in roots (37–56%), whereas starch was reduced. In
leaves, a marked increase in sucrose was detected from the first sampling onwards (36–66%), and the
same patter was observed for starch. Taken together, these results indicate that flooding altered carbon
and nitrogen pools and partitioning in citrus. On one hand, reduced nitrogen concentration appears to be a
consequence of impaired uptake and transport. On the other hand, the observed changes in carbohydrate
distribution suggest that translocation from leaves to roots was reduced, leading to significant starch
accumulation in leaves and further decreases in roots.
© 2012 Elsevier GmbH. All rights reserved.

Introduction physiological, biochemical, and molecular responses, most of which

Soil flooding is a major abiotic stress affecting large areas of
the world. Although variable in frequency and timing, waterlog-
ging modifies soil structure and nutrient dynamics in ecosystems
(Glazebrook and Robertson, 1999). In croplands, an excess of water
also transforms soil properties (Unger et al., 2009), usually lead-
ing to crop losses (Rao and Li, 2003; Schaffer et al., 2006). Many
species are sensitive, to a greater or lesser extent, and for sur-
vival in a flooded environment, have developed adaptations to
anaerobic conditions (Bailey-Serres and Voesenek, 2008). These
include acclimation mechanisms associated with a wide variety of
∗ Corresponding author. Tel.: +34 96 342 40 46; fax: +34 96 342 40 01.
E-mail address: iglesias dom@gva.es (D.J. Iglesias).
1
These authors contributed equally to this work.
lead to changes in anatomical or morphological traits (Colmer and
Voesenek, 2009).
The metabolic and physiological effects of flooding on plants
are derived mainly from a breakdown of the aerobic mitochon-
drial root respiration and subsequent oxygen deprivation. The
initial responses have been frequently revised in the literature
and include reductions in water flux from roots followed by
alterations in water use, nutrient uptake and dry weight parti-
tioning (Comstock, 2002; Colmer and Voesenek, 2009). At the leaf
level, the decline in stomatal conductance to avoid water, loss
together with the decrease in photosynthetic activity (Kozlowski,
1997), constitute a first barrier of defense against stress. Chem-
ical and hormonal factors have been identified as main players
in the signaling process, and changes in metabolite trafficking
and in the expression of stress-responsive genes have also been
reported (Dat et al., 2004). Further tissue damage and senescence

0176-1617/$ – see front matter © 2012 Elsevier GmbH. All rights reserved.
http://dx.doi.org/10.1016/j.jplph.2012.03.016
 B. Martínez-Alcántara et al. / Journal of Plant Physiology 169 (2012) 1150–1157
as well as a marked decrease in growth can finally lead to plant
death.
A general consequence of the alteration in root physiology is the
reduced uptake of elements, accountable for root to shoot dysfunc-
tions, changes in nutrient partitioning and, lastly, metabolic and
nutritional imbalances. Thus, endogenous concentrations of micro-
and macroelements can be radically modified under flooding con-
ditions, and multiple deficiencies usually appear (Pezeshki et al.,
1999), sometimes leading to severe carbon imbalances (Hsu et al.,
1999).
Although the response is variable among cultivars and species,
citrus is considered a flooding-sensitive crop, and the main effects
of inundation have been characterized in the last decades. Thus, the
effect on hydraulic processes in roots, together with an altered gas
exchange machinery in leaves (Vu and Yelenosky, 1991; Syvertsen
et al., 2007; García-Sanchez et al., 2007) have been classic objects of
study. Recent attention has also been paid to the signal transduction
mechanisms leading to these responses (Arbona and Gomez-
Cadenas, 2008; Rodríguez-Gamir et al., 2011). However, in spite
of the importance of carbon and nitrogen in determining many
physiological and agronomical traits (Iglesias et al., 2007), there is a
lack of knowledge regarding the influence of waterlogging on their
uptake, transport and distribution. Herein, we analyze the effects of
flooding on nitrogen and carbon resources and partitioning, paying
special attention to the carbohydrate changes in roots, leaves and
twigs. We also study the effects on nitrogen and carbon uptake and
translocation through the exogenous supplementation of labeled
nitrogen (15 N) and carbon (13 C) to roots and leaves, respectively.
Material and methods
Plant material and growth conditions
Ten-month-old plants of Carrizo citrange were used. Plants were
cultured under glasshouse conditions with supplementary light
(750 ␮mol m−2 s−1 ) to extend the photoperiod to 16 h. Tempera-
tures ranged between 16 and 18 ◦ C at night and 24 and 28 ◦ C by day,
and relative humidity was maintained between 65% and 80%. Dur-
ing the experimental period, photon flux density in the glasshouse
reached maximum values of 1250 ␮mol m−2 s−1 at 12:00 a.m. on
sunny days.
Plants were grown individually in 4 L pots filled with coarse
sand previously sterilized. 60 days before the beginning of the
experiment, all plants started to be irrigated twice per week
with the following nutrient solution: 3 mM Ca(NO3 )2 , 3 mM KNO3 ,
2 mM MgSO4 , 2.3 mM H3 PO4 , 17.9 ␮M Fe-EDDHA, 46.25 ␮M H3 BO3 ,
54.4 ␮M MnSO4 ·H2 O, 7.65 ␮M ZnSO4 ·7H2 O, 0.55 ␮M MoO3 and
0.5 ␮M CuSO4 ·5H2 O. The nutrient solution pH was adjusted to 6.5
with 1 M KOH. 1 L of solution per pot used in each watering event.
Excess solution drained out of the pot, thereby avoiding salt accu-
mulation in the sand.
One hundred sixty plants growing as a single shoot were
selected for uniformity of size at the beginning of the experiment.
Plants were then randomly divided into two uniform groups. One
group (control plants) was watered normally (as indicated above)
and well drained; the other was maintained under continuous sub-
strate waterlogging, which was administered by placing the pots
inside plastic containers in which the level of the nutrient solu-
tion remained 4 cm above the soil surface. N2 was bubbled into the
water to remove excess O2 and the nutrient solution was renewed
weekly. When required, an additional volume was supplemented to
maintain the water level. An opaque plastic sheet was used to cover
the surface of containers to avoid algal proliferation. The plants
were randomized over the experimental area and analyzed individ-
ually. A row of plants, not included in the experiment, was placed
around the perimeter.
1151
Flooding treatment was administered for 36 days, after which
individual seedlings were harvested and fractioned into leaves,
shoot and roots, fresh weighed and sampled on days 12, 24 and 36.
Additional samples were collected on days 6, 18 and 30 for carbohy-
drate analysis. Material was immediately frozen in liquid nitrogen
and freeze-dried for at least 96 h. Dried samples were weighed
and ground into fine powder using a laboratory ball mill (Retsch
MM301, Haan, Germany) and stored at −20 ◦ C until analysis.
Growth parameters
Growth rates were calculated according to Ruiz et al. (1997).
Relative growth rate (RGR) was considered as the increase in plant
DW per unit of plant DW per unit of time, and net assimilation rate
on a leaf weight basis (NARw ) was defined as the increase in plant
DW per unit of leaf DW per unit of time.
15 N labeling
Two methods for 15 N supplementation were used. First, a
pool of 36 plants (18 per treatment) was continuously supplied
with labeled K15 NO3 and Ca(15 NO3 )2 as nitrogen source (iso-
topic enrichment of 12% atom 15 N excess). On days 12, 24 and
36 after the experiment began, 6 plants per treatment were col-
lected and sampled as described above to determine differences
in the accumulation pattern and partitioning. In a complementary
experiment, another 36 plants were used for punctual “pulse” 15 N
supplementation assays to investigate time-related changes in the
uptake rate of the labeled nitrogen. In this case, series of 6 plants
per treatment were 15 N-supplemented for 4 days with the labeled
solution (isotopic enrichment of 30% atom 15 N excess) on days 8,
20 and 32. As above, samples for isotope analysis were collected on
days 12, 24 and 36.
13 CO pulse labeling and sampling
2
In order to trace the carbon translocation to different parts of
flooded and control plants, six plants per treatment were used for
13 C-isotope labeling 32 days after the beginning of the assay. The
portions of the seedlings aboveground were individually enclosed
in polyethylene bags (70 cm × 50 cm, resulting in a real chamber
volume of about 10 L) and exposed to a 13 CO2 -enriched atmo-
sphere. The 13 C was supplied by the injection of an air–gas mixture
containing 450 ppm of 13 CO2 , 20% N2 and O2 the remaining amount.
A device capable of delivering 10 s pulses every 5 min, injected the
air–gas mixture at a pressure of 1.5 bar within the bags. Isotope
pulse labeling was carried out under greenhouse conditions and
lasted for 4 h (between 09:00 and 13:00 h). A 3 mm diameter hole
allowed the gas mixture to leak to the outside air, thus avoiding
over-pressurization in the labeling bags. Just after labeling, bags
were opened and removed and leaves sampled to register the ini-
tial value (0 h) of ␦13 C. Four days after the isotope supplementation
(day 36), samples were collected as described above. In addition to
␦13 C, biomass specific 13 C accumulation in leaves, twigs and roots
was calculated as (13 C-organ/13 C-total plant)/(biomass organ/total
plant biomass) according to Bledsoe and Orians (2006).
Gas exchange parameters
Photosynthetic rate (ACO2 ) of single attached leaves was mea-
sured between 10:00 and 11:30 a.m. on sunny days, which
facilitated measurement under relatively stable conditions. Pho-
tosynthetically active radiation at the leaf surface was adjusted to
750 ␮mol photon flux density m−2 s−1 according to Iglesias et al.
(2002). A closed gas exchange CIRAS-2 (PP-systems, Hitchin, UK)
was used to take measurements. Leaf laminae were fully enclosed
1152 B. Martínez-Alcántara et al. / Journal of Plant Physiology 169 (2012) 1150–1157
within a PLC 6 (U) universal leaf autocuvette in a closed circuit
model and kept at 25 ± 0.5 ◦ C, with a leaf-to-air vapor deficit of
about 1.7 Pa. The air flow rate through the cuvette was 0.5 L min−1 .
Determinations were performed regularly on six plants per
treatment (Control, Flooding), using three uniform fully-expanded
mature leaves from the mid-stem zone. The average value of the
three leaves was considered as representative of each individual
plant.
Nitrogen, carbon, and isotope analysis
To determine total C (%), N (%) and the enrichments in 15 N and
13 C,samples were combusted in an element analyzer (NA 2500,
Thermo Finnigan, Bremen, Germany) coupled to an isotope ratio
mass spectrometer (Delta Plus, Thermo Finnigan) via a Finnigan
ConFloIII interface. Relative proportions of 15 N/14 N are expressed as
at.% 15 N excess which was calculated by subtracting natural abun-
dance of 15 N from 15 N enrichment of each sample. The natural
abundance of 15 N was considered to be the abundance of atmo-
spheric N2 0.3663 at.% according to the International Atomic Energy
Agency. Isotopic values of 13 C are expressed in ␦ units denoting
parts per thousand deviations (‰) relative to VPDB (Vienna Pee
Dee Bee Belemnite). All determinations were performed in dupli-
cate and after each set of ten samples, a 15 N or 13 C enriched standard
was run to ensure accuracy. The specific absorption rate of nitrogen
(SAR) was calculated as the variation in nitrogen amount per unit
of root DW per unit of time. The transport rate of this element from
root to shoot was calculated as the net increase in the amount in
the shoot per unit of root DW per unit of time (see Ruiz et al., 1997
and references therein).
Carbohydrate analyses
Samples for carbohydrate quantification were extracted with
80% ethanol, purified sequentially by cation and anion exchange
columns (Dowex 50X8-100 and 1X4-400, respectively, Sigma
Química, Madrid, Spain), and through C18 Sep-Pak cartridges
(Waters Chromatography, Barcelona, Spain). Sucrose, glucose and
fructose were quantified with a Waters HPLC system equipped
with a high performance carbohydrate column (4.6 mm × 250 mm,
Waters Chromatography, Barcelona, Spain) and a 2410 differen-
tial refractometer. Starch was determined in the remaining pellets
after extraction of soluble sugars. It was gelatinized by autoclav-
ing the samples, sodium-acetate buffer and amyloglucosidase were
added to the extracts, and enzymatic digestions performed for 2 h
at 55 ◦ C. After filtration, released glucose was estimated by HPLC as
described above.
Statistical analyses
Parameters were tested by Analyses of Variance (ANOVA) and
comparisons of means determined through the Least Significant
Differences (LSD) method at 95% confidence level. Statistics were
performed with Statgraphics Plus version 5.1 (Statistical Graphics,
Englewood Cliffs, NJ).
Results
Photosynthetic activity
Significant differences in photosynthetic rates were detected
between flooded and control plants throughout the course of the
treatment. Flooded plants exhibited lower values than controls
early on from the beginning of the experiments (Fig. 1). After 36
days of treatment, leaf photosynthesis of waterlogged plants was
Fig. 1. Photosynthetic rate (750 ␮mol photon flux density m−2 s−1 ) of fully
expanded leaves of Carrizo citrange seedlings during flooding treatment. Plants
were either non-stressed (control) or flooded. Data are mean values (n = 6 plants);
the error bars show standard errors and the different letters indicate statistically
significant differences (p ≤ 0.05) for the same sampling date.
severely affected, decreasing strongly compared to the controls
(more than 65%).
Plant biomass and dry weight partitioning
Fig. 2 shows the biomass – on a dry mass basis – of leaves,
twigs and roots produced per plant. At the end of the experiment,
flooding significantly reduced mean dry weight of the plants (7.27
vs. 6.45 mg DW plant−1 in control and flooded seedlings, respec-
tively). When biomass was studied in the different tissues, the data
revealed significant reductions in both leaf and root biomass. Thus,
by day 36, leaf biomass had decreased by 12.5%, although the values
were not significant, while the decline in root biomass was greater
than 30%. By this time, the root to shoot ratio had decreased about
30% in flooded compared to control seedlings (0.36 vs. 0.25, respec-
tively, p < 0.05, data not shown). When the growth parameters RGR
and NARw were examined (Fig. 3), the data revealed that the rela-
tive growth rate (RGR) of waterlogged plants markedly decreased
at the end of study (about 35%), parallel to a reduction (34%) in net
assimilation rate on a leaf dry weight basis (NARw ) at 24–36 days,
but not during the 0–12 and 12–24 day intervals.
Nitrogen content, distribution, uptake rate and transport
Total nitrogen content in the plant was strongly affected by
flooding from the first sampling date onwards (Fig. 4A). A signif-
icant reduction of total nitrogen in waterlogged plants appeared
12 days after the beginning of the assay and decreased progres-
sively with the time course relative to the control treatment. Thus,
by day 36, nitrogen content of intact control plants was 2.3-fold
that of the flooded seedlings. The distribution pattern of nitrogen
in the plant was also altered as a consequence of flooding (Fig. 4B).
Thus, by day 36, the nitrogen distribution in the flooded seedlings
was markedly affected: leaves significantly reduced their propor-
tion of nitrogen when compared to intact controls, whereas twigs
exhibited the opposite pattern. No significant effect was detected
in roots.
The trend in nitrogen concentration strongly paralleled nitrogen
content (Table 1). Whereas the decrease in nitrogen concentration
in leaves was observed from the first sampling date onwards, the
effect upon roots was apparent after 24 days of treatment. No effect
was observed on twigs. At the end of the study, reductions in nitro-
gen concentration ranged from 21% and 55% in roots and leaves,
respectively. No change in the proportion of proteic and soluble
nitrogen fractions was found (data not shown).
 B. Martínez-Alcántara et al. / Journal of Plant Physiology 169 (2012) 1150–1157 1153

Fig. 3. Effect of flooding treatment on relative growth rate (RGR) and net assimila-
tion rate on a leaf DW basis (NARw ). Data are mean values (n = 6 plants); the error
bars show standard errors and the different letters indicate statistically significant
differences (p ≤ 0.05) for the same sampling date.

Fig. 2. Evolution of dry weight biomass (g) of Carrizo citrange seedlings during flood-
ing treatment. Data are mean values (n = 6 plants); the error bars show standard
errors and the different letters indicate statistically significant differences (p ≤ 0.05)
for the same sampling date.

When nitrogen uptake and transport were estimated (Fig. 5A

and B), the data showed that the specific absorption rate on a
root DW basis was progressively reduced under flooding. At the
end of the study, values reached the greatest differences (35-fold).
The amount of nitrogen transported to shoots, also on a root DW

Table 1
Nitrogen concentration (%) of Carrizo citrange seedlings during flooding treatment.
Plants were either non-stressed (CT) or flooded (FL). Data are mean values (n = 6);
the error bars show standard errors and the different letters in each row indicate
statistically significant differences (p ≤ 0.05).

Days after treatment

0 12 24 36

CT FL CT FL CT FL CT FL

N (%)
Fig. 4. Total nitrogen content (panel A) and nitrogen relative distribution in leaves,
Leaves 2.08a 2.12a 3.10a 1.88b 3.13a 1.58c 3.05a 1.37d
twigs and roots (panel B) of Carrizo citrange Carrizo seedlings during flooding treat-
Twigs 0.87ab 0.76b 1.02a 0.94ab 1.03a 0.79b 1.15a 0.74b
ment. Plants were either non-stressed (control) or flooded. Data are mean values
Roots 2.16a 2.08a 2.12a 1.93ab 2.24a 1.83b 2.26a 1.78b
(n = 6); the error bars show standard errors and the different letters indicate statis-
Total plant 1.56ab 1.44b 1.95a 1.57ab 2.01a 1.30c 2.08a 1.16d
tically significant differences (p ≤ 0.05).
1154 B. Martínez-Alcántara et al. / Journal of Plant Physiology 169 (2012) 1150–1157
Fig. 5. Effect of flooding on nitrogen uptake and transport. (A) Specific nitrogen
absorption rate (SAR). (B) Transport root to shoot on a root DW basis. (C) Evolution
of 15 N uptake rate of Carrizo citrange seedlings during flooding treatment. Plants
were 15 N-supplemented for 4 days with the labeled solution (isotopic enrichment
of 30% atom 15 N excess) on days 8, 20 and 32. (D) Effect of flooding on 15 N rela-
tive distribution in leaves, twigs and roots of Carrizo citrange seedlings. From the
beginning of the experiment onwards, 15 N-treated plants were regularly supplied
with labeled K15 NO3 and Ca(15 NO3 )2 as nitrogen source (isotopic enrichment of 12%
atom 15 N excess) and samples were collected on day 36. In all cases, data are mean
values (n = 6); the error bars show standard errors and the different letters indicate
statistically significant differences (p ≤ 0.05).
basis, showed a similar tendency, and by day 36 values from con-
trols were more than 12-fold higher than those from waterlogged
plants. 15 N pulse supplementation assays proved a valuable tool
for precisely understanding how nitrogen uptake rate was affected
through the course of the experiment (Fig. 5C). Thus, whereas
no change was detected in control plants, flooding dramatically
decreased 15 N uptake rate from the first sampling date onwards
(reductions higher than 90% at the end of the experiment). When
expressed on a root DW basis, 15 N uptake rates (Fig. 5C) also fol-
lowed a similar, more remarked trend. Thus, by day 36, values
from flooded plants had drastically declined compared to control
seedlings (0.04 vs. 0.38 mg mg−1 day−1 , p < 0.05, data not shown).
Interestingly, the study of the relative distribution of the recovered
15 N (Fig. 5D) indicated that flooded leaves showed a strong decline
compared to controls (by 50%), whereas roots and twigs presented
Fig. 6. Effect of flooding on 13 C relative distribution (%) in leaves, twigs and roots of
Carrizo citrange seedlings. Plants were individually supplied with an air–gas mixture
containing 450 ppm 13 CO2 36 days after the beginning of the treatment and samples
were collected 4 days later. Data are mean values (n = 6 plants); the error bars show
standard errors and the different letters indicate statistically significant differences
(p ≤ 0.05) for the same organ.
the opposite trend, with significant increases ranging from 36 to
44%, respectively.
When the nutrient solution was continuously supplemented
with 15 N, the isotopic enrichment in the flooded seedlings was
found to be significantly reduced compared to intact control plants
(Table 2). The differences progressively increased throughout the
course of the experiment, and 36 days after treatment, the reduc-
tion was greater than 50% over the whole plant. Leaves showed the
greatest decline (nearly 70% compared to controls), and twigs and
roots also showed very significant effects (about 50% in both cases).
Carbon content and recovery of labeled 13 C
Carbon concentration did not vary significantly during the
experiment in the organs studied (43.5–45.7%), and no differences
were found between treatments (Table 3). However, leaf 13 CO2
assimilation was greatly influenced by flooding. At the end of the
study, the values of ␦13 C from flooded plants were strongly reduced
in both leaves and roots (22% and 37%, respectively). In addition,
waterlogging did induce a very perceptible change in the recovery
of 13 C from previously supplemented plants significantly, altering
the biomass-specific 13 C content in both organs.
13 CO labeling assays also demonstrated that the absorbed 13 C
2
was differentially distributed into the plant (Fig. 6). Thus, whereas
at the end of the experiment flooded leaves showed significantly
higher ␦13 C values than controls (26%), roots exhibited the opposite
pattern.
Carbohydrate content in roots and leaves
Since carbohydrates constitute the major source for carbon in
higher plants, we determined the content in roots and leaves of
both control and flooded citrus seedlings. Fig. 7 shows the car-
bohydrate concentration in roots throughout the study period. In
general, flooded seedlings showed a slight and progressive increase
in hexoses in roots. Significant differences were attained only at
the end of the study. In addition, sucrose concentration increased
significantly from the first sampling onwards (37–56%). By con-
trast, root starch content was reduced in flooded plants compared
to controls, and differences were markedly accentuated from day
18 onwards (18–35%).
In leaves, no change in hexose amount was detected under
flooding (Fig. 8), while a significant increase in sucrose concen-
tration was observed from day 12 onwards (36–66%). A marked
increase was also encountered when starch was studied (16–22%).
 B. Martínez-Alcántara et al. / Journal of Plant Physiology 169 (2012) 1150–1157 1155

Table 2
Atom 15 N excess (%) of Carrizo citrange seedlings during flooding treatment. Plants were either non-stressed (CT) or flooded (FL). From the beginning of the experiment
onwards, 15 N-treated plants were regularly supplied with labeled K15 NO3 and Ca(15 NO3 )2 as nitrogen source (isotopic enrichment of 12% atom 15 N excess). Data are mean
values (n = 6); the error bars show standard errors and the different letters in each row indicate statistically significant differences (p ≤ 0.05).

Days after treatment

0 12 24 36

CT FL CT FL CT FL CT FL

Atom 15 N excess (%)

Leaves – – 2.50c 0.65f 3.33b 0.94e 4.58a 1.53d
Twigs – – 2.35c 0.94e 3.39b 1.45d 4.43a 2.37c
Roots – – 2.79c 1.12f 3.94b 1.70e 4.88a 2.52d
Total plant – – 2.68c 0.93f 3.64b 1.35e 4.72a 2.19d

Table 3
Carbon concentration (%) and ␦13 C (‰) of Carrizo citrange seedlings determined 36 days after the beginning of the flooding treatment. Plants were either non-stressed (CT)
or flooded (FL). Data are mean values (n = 6); the error bars show standard errors and the different letters in each row indicate statistically significant differences (p ≤ 0.05).

Organ C concentration (%) ␦13 C (‰) Biomass specific 13 C content

CT FL CT FL CT FL

Leaves 43.77 43.51 257.70a 200.70b 0.97b 1.20a

Twigs 45.67 45.37 126.00a 128.80a 0.99a 0.98a
Roots 43.95 43.54 115.84a 73.18b 1.04a 0.79b

Fig. 7. Changes in carbohydrate content of roots of Citrange carrizo seedlings during Fig. 8. Changes in carbohydrate content of fully expanded leaves of Citrange car-
flooding treatment. Data are mean values (n = 6); the error bars show standard errors rizo seedlings during flooding treatment. Data are mean values (n = 6); the error
and the different letters indicate statistically significant differences (p ≤ 0.05) for the bars show standard errors and the different letters indicate statistically significant
same sampling date. differences (p ≤ 0.05) for the same sampling date.
1156 B. Martínez-Alcántara et al. / Journal of Plant Physiology 169 (2012) 1150–1157
Discussion
Our results illustrate how nitrogen and carbon pools of cit-
rus plants change in response to flooding conditions. Moreover,
through the study of the translocation of supplemented 15 N and 13 C
to roots and leaves, respectively, we demonstrate an impairment
of nitrogen and carbon incorporation, transport and accumulation
inside the plant.
The data presented here show that citrus seedlings exposed
to flooding progressively decreased their rates of photosynthe-
sis (Fig. 1). The effects on the gas exchange machinery, together
with other stomatal-derived responses, have been described in cit-
rus trees, and constitute early mechanisms preventing water loss
through the leaves (Vu and Yelenosky, 1991). Since photosynthesis
represents the major carbon source in higher plants, impaired CO2
assimilation as a consequence of stress should have harmful con-
sequences for carbon dynamics. In fact, although variable among
genotypes, the maintenance of stable photosynthetic performance
has previously been linked to flooding tolerance in citrus species
(Arbona et al., 2009).
As a consequence of anoxia, altered metabolism leads to a pro-
gressive loss in plant biomass and performance. Our results also
indicate that, although photosynthesis alteration was observed
early after the beginning of the treatment (Fig. 1), significant
decreases in relative growth rate and biomass were monitored
afterwards (Figs. 2 and 3). Thus, although flooded plants also
increased biomass progressively through the treatment, signifi-
cant differences between control and the flooding condition were
observed at the end of the experiment (day 36). The data suggest
that the decline in RGR was due to changes in NARw rather than
LAR, in agreement with an impaired photosynthetic activity and
the absence of differences in leaf dry weight between control and
flooded plants. The reduction of the net assimilation rate on a leaf
DW basis (NARw ) in flooded plants strongly paralleled the differ-
ences in relative growth rate (RGR), suggesting that the decline in
growth was associated, to a greater or lesser extent, to a photosyn-
thetic affection (Ruiz et al., 1997).
Typical responses under flooding conditions involve changes
in element uptake, as reviewed for various tropical and subtrop-
ical crops (Schaffer et al., 2006). In particular, nitrogen-associated
changes have usually been described to occur. However, vari-
ation in time and intensity has been not accurately assessed,
especially for woody plants. The data presented here confirm
that total plant nitrogen was significantly reduced in the flood-
ing environment, and N relative distribution was altered (Fig. 4A
and B). In field conditions, chemical properties of the sub-
strate can change with inundation (Unger et al., 2009) and
microorganism communities and enzymatic activities may also
be affected (Bossio and Scow, 1995; Eivazi, 1998), decreasing
nitrogen availability for growing plants (Ferreira et al., 2008).
However, in our experiment, the use of sand as substrate,
together with the regular renewal of the nutrient solution, min-
imized undesirable and artefactual reductions in the nitrogen
supply.
Although progressive through the course of the experiment, the
decline in nitrogen concentration and content was observed early
on after the treatment began, indicating a rapid induction of this
deficiency. It is not surprising that the progressive and general-
ized reduction and redistribution of nitrogen in all plant organs
finally altered growth and biomass production. In fact, the specific
absorption rate, an index of the uptake efficiency of roots, and trans-
port rate of nitrogen from root to shoot, showed significant, marked
reductions throughout the experiment (Fig. 5A and B). Neverthe-
less, our data did not reveal any changes in either the soluble or
the proteic nitrogen fraction during the study period. This absence
of redistribution of nitrogen fractions might be a consequence of a
readjustment, as suggested for other plant species (Komatsu et al.,
2010).
A powerful and unexplored line of evidence investigating the
direct effects on nitrogen uptake and transport under flooding con-
ditions is provided through experiments involving labeled isotopes.
In the present study, 15 N supplementation to the nutrient solu-
tion and further recovery was evaluated to determine the extent to
which nitrogen absorption and accumulation inside the plant was
affected. The data suggest that the decrease in total nitrogen was as
consequence, at least partially, of a significant reduction in nitro-
gen uptake rate (higher than 90% by day 36, Fig. 5C). Interestingly,
the absorbed 15 N was differentially transported and distributed
depending on the treatment. Thus, at the end of the experiment,
roots and twigs from flooded plants showed an increased recovery
of the labeled isotope –indicating accumulation- whereas leaves
showed the opposite pattern. These data not only appear to be asso-
ciated with the progressive reduction of the nitrogen pool at leaves,
but also suggest the possibility that the translocation of nitrogen
throughout the vascular system was notably impaired. Together
with a reduced demand from leaves, several physical, biochemi-
cal and molecular factors (changes in pH, transporter dysfunctions,
changes in aquaporin function, etc.) have been reported to con-
tribute to the impairment of the vascular loading and connections
under flooding (see Rodríguez-Gamir et al., 2011 and references
therein). Certain authors further indicate that many tree species
respond to this low nitrogen uptake and translocation rates by
changing organ characteristics (leaf area and/or number, root to
shoot ratio, root morphology, etc.), which might also contribute to
the changes observed in biomass.
The 13 CO2 labeling assay revealed a significant reduction in
the recovery of the labeled isotope in the waterlogged seedlings
(22–37%, Table 3). These parameters have been demonstrated to
be highly dependent on pool size and other metabolic processes,
so the observed decline does not necessarily imply reductions in
total carbon content and/or concentration in the plant (Wardlaw,
1990). In our experiment, although no differences were detected
in total carbon concentration through the course of the experi-
ment, total carbon content per plant was significantly decreased,
since flooded plants significantly decreased biomass (Fig. 2). Com-
pensation mechanisms that maintain stable carbon concentration
inside the plant like the observed reduction in growth and biomass,
together with the regulation of catabolic pathways (such as res-
piration rates) could also contribute to the observed responses
(Bailey-Serres and Voesenek, 2008).
Since carbohydrates constitute the main carbon reserves of
higher plants and maintenance of adequate carbon metabolism is a
prerequisite to surviving stress, the carbohydrate amounts and par-
titioning were analyzed in response to flooding. Like other species,
changes in carbohydrate concentration appear to be associated
with a reduction in growth and a ralentized use of energy sources.
Sucrose, glucose, fructose and starch are the major non-structural
sugars present in citrus leaves, and our data pointed to important
changes in their concentrations. Thus, Fig. 7 illustrates a signifi-
cant decline in the starch (the storage carbohydrate) concentration
in roots (35% by day 36), accompanied by a significant increase in
soluble sugars, principally sucrose (37–56%). Certain works pos-
tulate that the ability to utilize storage reserves like starch under
anoxia might be responsible for tissue survival under flooding con-
ditions (Goggin et al., 2007). Increased amounts of sucrose in the
root appear to be directly associated with starch hydrolization to
maximize energy supply and, simultaneously, to contribute to the
osmotic equilibrium of the cells. Furthermore, the early accumu-
lation of sucrose in flooded roots might also suggest a limited
ability to use assimilates, rather than lower supply. We also extend
the observation to the fact that carbohydrate availability did not
appear to be a limiting factor for citrus root metabolism under
 B. Martínez-Alcántara et al. / Journal of Plant Physiology 169 (2012) 1150–1157
flooding conditions, at least during the study period. This is con-
sistent with previous research indicating that soluble carbon,
abundant in flooded citrus roots, is a suitable indicator of short-
term physiological status under abiotic stresses (Gessler et al.,
2007), which could explain the survival of citrus plants in flooded
environments, at least during the initial stages.
In the aerial part of the plant, reduced leaf photosynthesis
was accompanied by increased sucrose and starch concentrations
(Fig. 8). It is not strange that the source limitation induced by the
decline in carbon assimilation led, to a greater or lesser extent,
to carbohydrate imbalances and further redistribution. Enhanced
non-structural carbohydrate storage in the aerial parts of many
plant species has been reported (Vu and Yelenosky, 1991; Gravatt
and Kirby, 1998; Kreuzwieser et al., 2004). The observed reductions
in dry weight under flooding, together with a decreased carbon
demand, are also probably associated with an increase in total non-
structural carbohydrate concentrations in leaves. Additionally, the
feed-back mechanism linking photosynthetic activity and carbohy-
drate content that operates in citrus plants (Iglesias et al., 2002)
could also contribute to the fine adjustment and partitioning of
leaf sugar pools to guarantee energy availability and enhance cell
survival.
Experiments involving assimilation of 13 CO2 through leaves
have also shed light on the regulation of leaf carbon uptake and
fate of carbohydrate during treatment. Impaired photosynthesis
coincided with an effective decrease in carbon assimilation, as con-
firmed by the reduced ␦13 C values obtained from both flooded
leaves and roots (Table 3), corroborating the source limitation
induced. Moreover, when 13 C relative distribution was examined
inside the plant (Fig. 6), the data revealed that leaves and roots were
differentially affected. Thus, 13 C recovery in leaves was increased
in waterlogged plants relative to controls, whereas roots showed
the opposite pattern. In this scenario, impaired carbon transloca-
tion from leaves to roots would also facilitate the observed starch
accumulation in leaves. Starch mobilization depends on numerous
factors and involves complex metabolic and molecular pathways
(Wardlaw, 1990). On one hand, changes in assimilate demand
under stress conditions have been demonstrated to alter carbon
partitioning according to source-sink compensation mechanisms,
leading to the observed mobilization of starch, the storage sugar
in citrus. The rate and potential appear to be dependent on com-
plex biochemical and molecular signaling pathways (Koch, 1996).
Amongst the enzymatic pathways affected, those involved in starch
degradation and sucrose metabolism have been object of extensive
study (Wardlaw, 1990 and references therein). Thus, together with
an altered demand for sucrose, physical changes in vascular prop-
erties as well as those in phloem loading and transport could also
contribute to the observed responses, as described in other species
(Saglio, 1985).
Taken together, these results indicate that flooding markedly
altered carbon and nitrogen pools and partitioning in citrus. On
one hand, reduced nitrogen concentration in the plant appears to
be a consequence of impaired uptake and transport. On the other,
the observed changes in carbohydrate distribution suggest that
translocation from leaves to roots was reduced, leading to signifi-
cant starch accumulation in leaves and further decrease in roots.
Acknowledgements
This work was funded by the Instituto Nacional de Investigación
y Tecnología Agraria y Alimentaria (RTA2008-00060), Generalitat
Valenciana and FEDER funds.
1157
References
Arbona V, Gomez-Cadenas A. Hormonal modulation of citrus responses to flooding.
J Plant Growth Regul 2008;27:241–50.
Arbona V, Lopez-Climent MF, Perez-Clemente RM, Gomez-Cadenas A. Maintenance
of a high photosynthetic performance is linked to flooding tolerance in citrus.
Environ Exp Bot 2009;66:135–42.
Bailey-Serres J, Voesenek L. Flooding stress: acclimations and genetic diversity. Ann
Rev Plant Biol 2008;59:313–39.
Bledsoe TM, Orians CM. Vascular pathways constrain 13 C accumulation in
large root sinks of Lycopersicon Esculentum (Solanaceae). Am J Bot 2006;93:
884–90.
Bossio DA, Scow KM. Impact of carbon and flooding on the metabolic diver-
sity of microbial communities in soils. Appl Environ Microbiol 1995;61:
4043–50.
Colmer TD, Voesenek LACJ. Flooding tolerance: suites of plant traits in variable
environments. Funct Plant Biol 2009;36:665–81.
Comstock JP. Hydraulic and chemical signalling in the control of stomatal conduc-
tance and transpiration. J Exp Bot 2002;53:195–200.
Dat JF, Capelli N, Folzer H, Bourgeade P, Badot PM. Sensing and signalling during
plant flooding. Plant Physiol Biochem 2004;42:273–82.
Eivazi F. Urease activity in soils subjected to flooding and its effect on nitrogen uptake
by corn and wheat. J Plant Nutr 1998;21:699–705.
Ferreira JL, Coelho C, Magalhaes PC, Sant’ana GC, Borem A. Evaluation of min-
eral content in maize under flooding. Crop Breed Appl Biotech 2008;8:
134–40.
García-Sanchez F, Syvertsen JP, Gimeno V, Botia P, Perez-Perez JG. Responses to
flooding and drought stress by two citrus rootstock seedlings with different
water-use efficiency. Physiol Plant 2007;130:532–42.
Gessler A, Keitel C, Kreuzwieser J, Matyssek R, Seiler W, Rennenberg A. Poten-
tial risks for European beech (Fagus sylvatica L.) in a changing climate. Trees
2007;21:1–11.
Glazebrook HS, Robertson AI. The effect of flooding and flood timing on leaf litter
breakdown rates and nutrient dynamics in a river red gum (Eucalyptus camald-
ulensis) forest. Austr J Ecol 1999;24:625–35.
Goggin DE, Colmer TD. Wheat genotypes show contrasting abilities to recover
from anoxia in spite of similar anoxic carbohydrate metabolism. J Plant Physiol
2007;164:1605–11.
Gravatt DA, Kirby CJ. Patterns of photosynthesis and starch allocation in seedlings
of four bottomland hardwood tree species subjected to flooding. Tree Physiol
1998;18:411–7.
Hsu YM, Tseng MJ, Lin CH. The fluctuation of carbohydrates and nitrogen compounds
in flooded wax-apple trees. Bot Bull Acad Sin 1999;40:193–8.
Iglesias DJ, Lliso I, Tadeo FR, Talón M. Regulation of photosynthesis through source:
sink imbalance in citrus is mediated by carbohydrate content in leaves. Physiol
Plant 2002;116:563–72.
Iglesias DJ, Cercós M, Colmenero-Flores JM, Naranjo MA, Ríos G, Naranjo MA, et al.
Physiology of citrus fruiting. Brazil J Plant Phys 2007;19:333–62.
Koch KE. Carbohydrate-modulated gene expression in plants. Annu Rev Plant Physiol
Plant Mol Biol 1996;47:509–40.
Komatsu S, Sugimoto T, Hoshino T, Nanjo Y, Furukawa K. Identification of flooding
stress responsible cascades in root and hypocotyl of soybean using proteome
analysis. Amino Acids 2010;38:729–38.
Kozlowski TT. Responses of woody plants to flooding and salinity. Tree Physiol
Monogr 1997;1:1–29.
Kreuzwieser J, Papadopoulou E, Rennenberg H. Interaction of flooding with carbon
metabolism of forest trees. Plant Biol 2004;6:299–306.
Pezeshki SR, DeLaune RD, Anderson PH. Effect of flooding on elemental uptake and
biomass allocation in seedlings of three bottomland tree species. J Plant Nutr
1999;22:1481–94.
Rao R, Li YC. Management of flooding effects on growth of vegetable and selected
field crops. HortTech 2003;13:610–6.
Rodríguez-Gamir J, Ancillo G, González-Mas MC, Primo-Millo E, Iglesias DJ, Forner-
Giner MA. Root signalling and modulation of stomatal closure in flooded citrus
seedlings. Plant Physiol Biochem 2011;49:636–45.
Ruiz D, Martínez V, Cerdá A. Citrus response to salinity: growth and nutrient uptake.
Tree Physiol 1997;17:141–50.
Saglio PH. Effect of path or sink anoxia on sugar translocation in roots of maize
seedlings. Plant Physiol 1985;77:285–90.
Schaffer B, Davies FS, Crane JH. Responses of subtropical and tropical fruit trees to
flooding in calcareous soil. HortSci 2006;41:549–55.
Syvertsen JP, Gimeno V, García-Sánchez F, Botía P. Responses to flooding and drought
stress by two citrus rootstock seedlings with different water-use efficiency.
Physiol Plant 2007;130:532–42.
Unger IM, Motavalli PP, Muzika RM. Changes in soil chemical properties with flood-
ing: a field laboratory approach. Agric Ecosyst Environ 2009;131:105–10.
Vu J, Yelenosky G. Photosynthetic responses of citrus trees to soil flooding. Physiol
Plant 1991;81:7–14.
Wardlaw IF. Tansley Review No. 27 The control of carbon partitioning in plants. New
Phytol 1990;116:341–81.