Physiologia Plantarum 0: 0–0.

2020 © 2020 Scandinavian Plant Physiology Society, ISSN 0031-9317

Physiological and metabolomic responses of bermudagrass

(Cynodon dactylon) to alkali stress
Tiantian Yea,b, Yanping Wanga, Yu-Qi Fengb and Zhulong Chana,c,*
a
Key Laboratory of Horticultural Plant Biology, Ministry of Education; College of Horticulture and Forestry Sciences, Huazhong Agricultural University,
Wuhan, Hubei, 430070, China
b
Key Laboratory of Analytical Chemistry for Biology and Medicine (Ministry of Education), Department of Chemistry, Wuhan University, Wuhan, 430072,
China
c
College of Life Science and Technology, Hubei Engineering University, Xiaogan, Hubei 432000, China

Correspondence Bermudagrass (Cynodon dactylon) is a widely used warm-season turfgrass

*Corresponding author,
species with superior stress tolerance except for cold. In this study, a com-
e-mail: zlchan@mail.hzau.edu.cn
parative analysis of the responses to alkali stress in bermudagrass at the
Received 15 June 2020; physiological and metabolomic levels were performed. Mild alkali with
revised 29 August 2020 relatively low pH slightly inhibited growth of bermudagrass as evidenced
by lower electrolyte leakage, more rapid growth and higher survival rate
doi:10.1111/ppl.13209
when compared to moderate and severe alkali treatments. Moreover, the
amount of 37 metabolites including amino acids, organic acids, sugars
and sugar alcohols were modulated by the alkali treatments. Among them,
15 metabolites were involved in carbon and amino acid metabolic path-
ways. Under mild alkali stress, bermudagrass possibly slowed down
metabolisms to maintain basic growth. However, moderate and severe
alkali-stressed plants accumulated significantly higher amount of carbohy-
drates which might result in carbon starvation. Taken together, alkali stress
had severely inhibitory effect partially due to combined ionic stress and
high pH stress. These results suggested that bermudagrass employed differ-
ent strategies in response to alkali stresses with different pH and ionic
values.

Introduction
and ionic signaling perception and adjustment
Alkalization and salinization have been considered as (Zhu 2002). However, studies on alkali stress have rarely
major environmental threats to agricultural systems, been performed. Due to the high pH, alkaline salts
which severely affect the growth, physiology and metab- (NaHCO3 and/or Na2CO3) have more destructive and
olism of agricultural crops and grasses (Lunde et al. 2007, complex effects on plants than that of salinization caused
Hu et al. 2015). Thus far, plant responses to salinity stress by neutral salts (NaCl and/or Na2SO4) (Yang et al. 2009,
caused by NaCl have been extensively studied (Shi Hu et al. 2015, Chen et al. 2018). Alkali stress exhibits
et al. 2013, Wu et al. 2013, Ye et al. 2016, Teng the same stress factors (ionic and osmotic stresses) as neu-
et al. 2018). When plant encounte NaCl stress, they tral salt stress, but becomes aggravated when combined
maintain osmotic and ion homeostasis via rapid osmotic with high pH stress (Yang et al. 2009, Guo et al. 2015).

Abbreviations – CAT, catalase; DREPP, plasma membrane polypeptide; EL, electrolyte leakage; GC, gas chromatograph; GPX,
glutathione peroxidase; GR, glutathione reductase; GSH, glutathione; GST, glutathione S-transferase; MDA, malondialdehyde;
MS, mass spectrometer; POD, peroxidase; ROS, reactive oxygen species; SOD, superoxide dismutase; TCA, tricarboxylic acid.

Physiol. Plant. 0, 2020 1

The high pH surrounding the plant roots cause proton and comparative metabolomics approach was applied
scarcity and disrupts the cross membrane potential, to investigate the mechanisms of bermudagrass in
inhibit the absorption of majority of ions that alter the response to different levels of alkali stresses.
availability of nutrients, and disrupt the balance of ions
(Yang et al. 2009). Recently, Hu et al. (2018) screened
the tolerance of 30 switchgrass lines to alkali stress and Materials and methods
found that the activity of catalase (CAT) can be consid- Plant materials and growth conditions
ered as a key physiological parameter for evaluation of
switchgrass tolerance to alkali stress. Growth of Sporobo- Bermudagrass seeds of the Yukon variety were first verna-
lus virginicus was significantly inhibited by alkali stress lized for 4 days in deionized water at 4 C in darkness,
and the gene encoding plasma membrane polypeptide then sterilized with 70% alcohol for 5 min and sown in
1 (DREPP1) was found to be involved in the alkali stress plastic pots filled with vermiculite. The growth room
responses (Theerawitaya et al. 2018). Under alkali stress, was controlled at 25  2 C, with an irradiance of about
the SvDREPP1 mRNA expression was upregulated by 3.5 150 μmol quanta m−2 s−1, 16/8 h light/dark cycles, and
times in the leaves and 6.0 times in the root compared to about 65% relative humidity. The plants were watered
the control (Theerawitaya et al. 2018). However, the every other day to 100% field capacity and irrigated
response and adaptation mechanisms to alkali stress in twice every week with 50 ml of half-strength Hoagland’s
grass species aremains elusive. nutrient solution. The 14-day-old seedlings of nearly the
Metabolomics is a powerful tool to elucidate plant same size were subjected to alkali stress treatment.
metabolic responses to abiotic stresses (Obata and Fer-
nie 2012). It focuses on a global profile of the low molec- Experimental design of stress treatments
ular weight metabolites which are the end products,
initiators or intermediates of important physiological For control treatment, plants were watered daily with the
and biochemical reactions (Kumari and Parida 2018). nutrient solution (pH 6.5). For alkali stress treatment, the
The presence and relative concentrations of metabolites alkaline solution was prepared as described by Zhang
can provide information on metabolic status in an organ- et al. (2012) and Hu et al. (2015) with slight modifica-
ism to assess the biological responses induced by tions. Briefly, three different pH values was obtained
exogenous factors (Kumari et al. 2015, Kumari and Par- through mixing 100 mM NaHCO3 and Na2CO3 at differ-
ida 2018). These metabolites may be key protective sub- ent ratios, including mild (pH 8.0, 100% NaHCO3,
stances in plant response to abiotic stress. Therefore, the 100 mM Na+); moderate (pH 9.6, 90% NaHCO3:10%
metabolic profile of plants provide a new approach to Na2CO3, 110 mM Na+) and severe (pH 10.6, 50%
identify compounds that play key roles during plant stress NaHCO3:50% Na2CO3, 150 mM Na+). The desired pH
response and reveal new pathways associated with stress value was adjusted with NaOH or acetic acid
tolerance (Kumari and Parida 2018). (Table S1). All bermudagrass plants were stressed for
Bermudagrass (Cynodon dactylon) is one of the most 21 days and the survival rate was recorded after 7 days
important warm-season turfgrasses and is widely used rewatering. Leaf samples were collected at 0, 7, 14,
in the construction of golf courses, lawns, sport fields 21 days under both control and stress conditions for anal-
and wetland vegetation restoration (Shi et al. 2012, ysis of physiological parameters. Since the electrolyte
2013, Ye et al. 2015). Changes in water status, antioxi- leakages (ELs) of moderate alkali-treated plants at
dant defense system and osmolyte accumulation might 14 days reached nearly 50%, the leaf samples after mid-
be contributed to the salt tolerance of bermudagrass vari- term treatment (14 days) were harvested for metabolite
eties (Ye et al. 2016). Exogenous application of small assay.
molecules increased salt stress tolerance of C. dactylon
(Shi et al. 2013, 2015). Moreover, the function of a num-
Determination of electrolyte leakage
ber of abiotic stress-responsive genes from bermudagrass
has been characterized. Overexpression of a C. dactylon For EL assay, detached plant leaves (about 0.2 g) were
stress-responsive nuclear factor Y gene (Cdt-NF-YC1) in gently shaken in deionized water at room temperature
rice resulted in increased tolerance to salt as well as for 6 h at 150 rpm to determine the initial conductivity.
increased sensitivity to abscisic acid (Chen et al. 2015). Leaf samples were then boiled at 121 C for 20 min and
To date, limited information is available on the cooled to room temperature to determine the fully
responses of C. dactylon to alkali condition. The detailed divided conductivity. The EL was calculated as the ratio
metabolomic changes in C. dactylon in response to alkali of the initial conductivity to fully releasing conductivity
stress are largely unknown. In this study, physiological (Shi et al. 2012).

2 Physiol. Plant. 0, 2020

Measurement of proline content extracted in 1.4 ml of 100% methanol (precooled at
−20 C). Ribitol (60 μl, 0.2 mg ml−1) was added to each
Proline content was estimated using the method descibed
sample as an internal quantitative standard. Then the
in Shi et al. (2012) with known concentration of L-proline
samples were shaken for 10 min at 70 C in a thermo-
as standard. Briefly, 0.25 g leaf samples were ground and
mixer. After centrifugation at 11 000 g for 10 min, the
extracted in 3% (w/v) sulfosalicylic acid at 100 C for
supernatant was transferred to a new tube. Then,
10 min. Then, 2 ml glacial acetic acid and 2 ml ninhydrin
750 μl chloroform and 1.4 ml Milli-Q H2O were added
reagent were added. The mixed solution was boiled at
to the sample. The well-mixed sample was centrifuged
100 C for 40 min, cooled to room temperature and the
at 2200 g for 15 min, and 150 μl supernatant was trans-
absorbance measured at 520 nm. Finally, the proline
ferred into a 1.5 ml tube. The extracted metabolites were
level was calculated according to the standard curve
derivatized with methoxyamination reagent, and
formed by known concentration of L-proline.
subsequent trimethylsilylated with N-methyl-N-
Quantification of sucrose and soluble total sugars trimethylsilyltrifluoroacetamide (MSTFA). For the deriv-
atization, an empty reaction tube was prepared as a
The sucrose content was determined by measuring the blank control, and also a C10, C12, C15, C19, C22,
absorbance at 480 nm, and the soluble total sugar con- C28, C32 and C36 n-alkane mixtures were used for the
tent of samples was examined by measuring the absor- determination of RIs to improve the accuracy of
bance at 620 nm. The levels of sucrose and soluble qualitation.
total sugar were calculated using the standard curve with
known concentration of sucrose and glucose, respec- Instruments and analytical conditions
tively (Shi et al. 2012).
Analysis of metabolites was performed on an Agilent
7890 gas chromatograph (GC) with a model 5975 mass
Determination of malondialdehyde content
spectrometer (MS) (Agilent Technologies, USA) accord-
The malondialdehyde (MDA) content in the samples ing to the procedure of Shi et al. (2014c). The separation
from the control and stress treatments was extracted using of metabolites was achieved on DB-5MS capillary
thiobarbituric acid reagent and the MDA concentrations (30 × 0.25 × 0.25 mm, Agilent J&W GC Column,
were determined at 450, 532 and 600 nm of absorbance USA). One microliter of derivatizated sample was
with a spectrometer as described previously (Shi injected for analysis. The measurement was performed
et al. 2012). with electron impact ionization (70 eV) in the full scan
mode (m/z from 30 to 600). The inlet temperature was
Determination of reactive oxygen species and set at 260 C. After a 6-min solvent delay, the initial GC
antioxidant enzyme activities oven temperature was set at 60 C; after injection for
The protein content was quantified using Bradford 1 min, the GC oven temperature was raised to 280 C with
method (Bradford 1976). The content of H2O2 was mea- 15 C min−1, and held at 280 C for 15 min finally. The
sured using the titanium sulfate regent as described by Shi injection temperature was set to 280 C and ion source
et al. (2012). For the O2•− content assay, a plant O2•− temperature was adjusted to 230 C. Helium was used
ELISA Kit (10-40-488, China) was used, and the absor- as the carrier gas with a constant flow rate set at
bance was quantified at 405 nm. 1 ml min−1.
The antioxidant enzyme activities including glutathi-
Metabolite data processing and analysis
one reductase (GR, EC 1.6.4.2), superoxide dismutase
(SOD, EC 1.15.1.1) and glutathione peroxidase (GPX, The Agilent MSD Productivity Chemstation software
EC 1.11.1.9) activities were determined using and GR were used for peak finding, peak integration and reten-
Assay Kit (S0055, Beyotime, China), Total SOD Assay tion time correction. The metabolites were identified
Kit with WST-1 (S0102, Beyotime), Total GPX Assay Kit based on retention time index specific masses, via com-
(S0058, Beyotime), respectively, according to the manu- paring with reference spectra in mass spectral libraries
facturer’s instructions. (NIST 2005, Wiley 7.0). Three biological replicates and
three technical replicates were used for the metabolo-
Metabolites extraction and derivatization mics analysis.
The metabolites extraction and derivatization were per-
Cluster analyses
formed as described by Lisec et al. (2015) and Ye
et al. (2016). Briefly, 100 mg of fresh plant leaves were Cluster program (http://bonsai.ims.u-tokyo.ac.jp/~mde
ground into a fine powder with liquid nitrogen, and hoon/software/cluster/) was used for hierarchical cluster

Physiol. Plant. 0, 2020 3

Fig. 1. Effects of alkali on growth of bermudagrass. (A) The photograph showing bermudagrass that subjected to control and stress conditions for
21 days. (B) Survival rate of bermudagrass after 21 days of control and stress treatments. (C) Electrolyte leakage; (D) shoot length and (E) root length.
Two-weeks-old bermudagrass were subjected to stress treatments. The data represent the means of three independent experiment  SE (n = 3), and
data followed by different letters are significantly different from each other at P < 0.05 according to Duncan’s method. Mild: pH 8.0, 100% NaHCO3,
100 mM Na+; moderate: pH 9.6, 90% NaHCO3 + 10% Na2CO3, 110 mM Na+; severe: pH 10.6, 50% NaHCO3 + 50% Na2CO3, 150 mM Na+.

analysis and JAVA TREEVIEW (http://jtreeview.sourceforge. For each independent experiment, at least 35 bermuda-
net/) was used to display the resulting tree figures grass plants were collected for sample preparation.
(de Hoon et al. 2004).
Results
Effect of alkali stress on growth of bermudagrass
Statistical analysis
Bermudagrass seedlings under control condition grew
All the experiments in this study were performed three well with normally elongated shoot and root length
times. Treatments were arranged as a completely ran- (Fig. 1A). Alkali stress severely inhibited the growth of
domized design with three replicates. The results were bermudagrass, even under mild (pH 8.0, 100%
shown as means  SE (n = 3). SPSS 13.0 software was used NaHCO3, 100 mM Na+) alkali stress condition (Fig. 1A).
for statistical analysis. A one-way analyses of variance Moderate and severe alkali stress conditions significantly
(ANOVA) followed by Duncan’s multiple range test was reduced the survival rates (Fig. 1B). After 21 days of treat-
employed to assess the statistical significance of the treat- ments, less than 46 and 16% of the seedlings survived
ment differences. Different letters above the columns in under moderate and severe alkali treatments, respec-
each figure indicate significant differences at *P < 0.05. tively, while all the seedlings survived under mild alkali

4 Physiol. Plant. 0, 2020

 damages induced by abiotic stress and play a key role
in plant redox metabolism (Couée et al. 2006). In this
study, mild alkali conditions had no significant effect on
proline content; however, moderate and severe alkali
stress treatments increased proline content at 14 and/or
21 days (Fig. 2A). In addition, soluble sugar and sucrose
contents showed similar changes as proline after alkali
stress treatment (Fig. 2B,C). These results indicated that
mild alkali stress did not affect the production of compat-
ible solutes in bermudagrass, but long-term moderate
and severe alkali stress significantly increased the con-
tent of compatible solutes.

Modulation of ROS metabolism in bermudagrass

after alkali treatment
Alkali stress increased the H2O2 content in bermudagrass
after long-term treatment (21 days) (Fig. 3A). Moderate
and severe alkali-treated plants showed significantly
higher H2O2 than in control and mild alkali-treated plants
(Fig. 3A). Mild alkali treatment did not change O2•− con-
tent and only slightly changed the MDA content (Fig. 3B,
C). However, after severe alkali treatment, MDA content
significantly increased after 7 days, and the O2•− content
increased after 21 days of treatment (Fig. 3B,C). Glutathi-
one (GSH) content showed no significant changes after
mild alkali stress but significant increase was seen after
severe alkali treatment for 21 days (Fig. 3D).
The activities of GR, GPX and SOD had similar
Fig. 2. Osmolytes accumulation in bermudagrass under alkali condition.
changes and were progressively induced with prolonged
(A) Proline content; (B) soluble sugars and (C) sucrose content of
bermudagrass. Two-weeks-old bermudagrass were subjected to stress alkali treatments (Fig. 4A-C). Increase of alkali pH value
treatments. The results shown are means  SE (n = 3), and the results and Na+ resulted in significantly enhanced GR, GPX
followed by different letters are significantly different from each other and SOD activities. These results indicated that alkali
at P < 0.05 according to Duncan’s method. Mild: pH 8.0, 100% stress modulated antioxidant enzyme activities to detox-
NaHCO3, 100 mM Na+; moderate: pH 9.6, 90% NaHCO3 + 10%
Na2CO3, 110 mM Na+; severe: pH 10.6, 50% NaHCO3 + 50% Na2CO3,
ify reactive oxygen species (ROS).
150 mM Na+.
Correlation analysis of physiological data in
bermudagrass after alkali treatment
conditions (Fig. 1B). EL progressively increased with the
duration of alkali stress treatment (Fig. 1C). However, To further dissect the relationship among different physi-
severe alkali treatment caused significant damages to ological parameters, we performed cluster analysis of the
the plants evidenced by nearly 90% EL (Fig. 1C). The investigated physiological data. The results showed that
shoot and root length of alkali-treated bermudagrass sig- the data could be divided into two clusters (Fig. 5A). Root
nificantly declined when compared to control plants at length, shoot length and survival rate decreased follow-
21 days after treatment (Fig. 1D,E). These results indi- ing the increment of alkali concentration, indicating that
cated that moderate and severe alkali treatments caused alkali treatment inhibited growth of bermudagrass plants.
severe cell membrane damages and greatly inhibited ber- Meanwhile, all other physiological data increased from
mudagrass growth, while the mild alkali had less inhibi- mild to severe alkali concentrations (Fig. 5A), including
tory effect on the growth of bermudagrass. EL, MDA, sugar, ROS, proline and antioxidant enzyme
activity. Correlation analysis showed that EL, MDA and
H2O2 contents had strong correlation with other parame-
Effect of alkali on compatible solute accumulation
ters as evidenced by R2 > 0.8, while SOD activity exhib-
Compatible solutes, including soluble sugars and pro- ited relatively weak correlation with proline, sugar,
line, protect the macromolecular structure from the sucrose and GSH contents with R2 < 0.7 (Fig. 5B).

Physiol. Plant. 0, 2020 5

 Fig. 4. Effects of alkali stress on activity of enzymatic antioxidants in
bermudagrass. (A) GR; (B) GPX and (C) SOD activities. Two-weeks-old
bermudagrass were subjected to stress treatments. The relative
activities were quantified as fold change in comparison with
bermudagrass under unstress condition. The data represent the means
of three independent experiment  SE (n = 3), and data followed by
different letters are significantly different from each other at P < 0.05
according to Duncan’s method. Mild: pH 8.0, 100% NaHCO3, 100 mM
Na+; moderate: pH 9.6, 90% NaHCO3 + 10% Na2CO3, 110 mM Na+;
severe: pH 10.6, 50% NaHCO3 + 50% Na2CO3, 150 mM Na+.

Fig. 3. Change of ROS and non-enzymatic antioxidant level of

bermudagrass under alkali stress condition. (A) H2O2 content; (B) O2•− performed in bermudagrass after 14 days alkali stress
content; (C) MDA level and (D) GSH content. Two-weeks-old treatment. In total, 37 metabolites, including 8 amino
bermudagrass were subjected to stress treatments. The data represent the
acids, 15 sugars, 9 organic acids, 2 sugar alcohols, 2 fatty
means of three independent experiment  SE (n = 3), and data followed
by different letters are significantly different from each other at P < 0.05 acids and 1 other (phosphoric acid) were detected (Fig. 6,
according to Duncan’s method. Mild: pH 8.0, 100% NaHCO3, 100 mM Table S2). The data showed that the contents of 21 and 5
Na+; moderate: pH 9.6, 90% NaHCO3 + 10% Na2CO3, 110 mM Na+; metabolites were significantly increased and decreased,
severe: pH 10.6, 50% NaHCO3 + 50% Na2CO3, 150 mM Na+. respectively after severe alklai stress. Mild alkali treat-
ment decreased, while moderate and severe alkali stres-
ses increased the amount of most metabolites. Contents
Identification of metabolites changed by alkali
of all detected amino acid increased significantly under
stress
moderate and severe alkali stress conditions, while the
Metabolomic analyses based on Gas chromatography levels of alanine, glycine, proline, serine and threonine
time-of-flight mass spectrometry (GC-TOF-MS) were showed no significant changes after mild alkali

6 Physiol. Plant. 0, 2020

Fig. 5. Cluster and correlation analyses of physiological data in bermudagrass. (A) Cluster analysis of physiological data after alkali treatment; Log2-based
values of physiological data at 21 days after alkali treatment were used for cluster analysis. (B) Correlation analysis of physiological data at 21 days after
alkali treatment. R2 values were presented.

treatment. Interestingly, the contents of disaccharides Carbon and amino acid pathways affected by alkali
(sucrose, lactose, lactulose, gentiobiose and maltose) sig- stress
nificantly increased after moderate and severe alkali
Among the 37 metabolites detected, 15 of them are
treatments but only slightly increased or decreased under
involved in the carbon and amino acid metabolic path-
mild alkali stress condition. Meanwhile, the content of ways (Fig. 7). Valine and alanine derived from pyruvate,
monosaccharides (fructose, allose, glucose, galactose, glycine and serine derived from 3-phosphoglycerate,
mannose and talose) declined significantly by mild alkali and proline and asparagine derived from tricarboxylic
treatment but declined slightly by moderate and severe acid (TCA) increased in bermudagrass under both moder-
alkali stresses. The content of most organic acids and fatty ate and severe alkali stress conditions (Fig. 7). Pyruvate-
acids significantly increased upon severe alkali treat- derived alanine, TCA-derived amino acids such as pro-
ment, but slightly increased by mild alkali (Fig. 6, line and asparagine increased only after moderate and
Table S2). severe alkali treatments (Fig. 7). Meanwhile, the

Physiol. Plant. 0, 2020 7

Fig. 6. Effects of alkali stresses on metabolites in bermudagrass. Two-weeks-old bermudagrass were subjected to stress treatments and metabolites were
extracted after 14-days treatment. Hierarchical cluster analysis of 37 compounds affected by salt and alkali stress conditions in bermudagrass using
CLUSTER 3.0 software package. The resulting tree figure was obtaining using the JAVA TREEVIEW. The absolute content of all metabolites was listed as
Table S2. Mild: pH 8.0, 100% NaHCO3, 100 mM Na+; moderate: pH 9.6, 90% NaHCO3 + 10% Na2CO3, 110 mM Na+; severe: pH 10.6, 50%
NaHCO3 + 50% Na2CO3, 150 mM Na+.

abundance of the disaccharide sucrose increased under bermudagrass variety ‘Yukon’ after 21 days of 400 mM
alkali stress (Fig. 7). NaCl treatment (Ye et al. 2016). These results indicated
that severe alkali stress may have more severe destructive
effects on bermudagrass than that of salt stress, which
Discussion
may be due to the combination of high pH stress and
Alkali stress greatly inhibits plant growth, development ionic stress (Yang et al. 2009, Guo et al. 2010, 2015,
and production. In plant cells, alkali condition leads to Hu et al. 2015). Two weeks of alkali stress treatment with
cell membrane damages. EL is a key parameter to assess 10 mM Na+ at pH 9.8 severely inhibited germination and
cell membrane stability in response to abiotic stress. In radicle elongation in alfalfa. Alfalfa plants did not survive
this study, alkali treatment caused significantly increased after treatment with 50 mM Na+ at pH 10.1 (Li et al. 2010).
EL and membrane damage in bermudagrass (Fig. 1B). Bermudagrass treated for 3 weeks with alkali stress
Previously we observed slight increase of EL in the showed slightly inhibited growth under mild alkali

8 Physiol. Plant. 0, 2020

Fig. 7. Functions of alkali modulated metabolites in the carbon metabolic pathway. A total of 15 metabolites from 37 assayed metabolites were assigned
to the carbon metabolic pathway. The detailed levels of these metabolites were shown in Table S2. The assigned metabolites were exhibited as boxes
with different colors. Red, significant increase; blue, significant decrease; white, no significant change.

condition with 100 mM Na+ at pH 8.0, and 15% survival endogenous ROS homeostasis, plants have evolved effi-
rate after treatment with 150 mM alkali stress at pH 10.6, cient enzymatic and non-enzymatic antioxidative sys-
indicating bermudagrass was relatively tolerant to alkali tems to protect themselves against oxidative damage
stress. and fine modulation of low levels of ROS for signaling
Abiotic stresses lead to overproduction of ROS and transduction. Both enzymatic antioxidants, like SOD,
oxidative stress in the plant (Pastori and Foyer 2002, CAT, peroxidase (POD), GR, dehydroascorbate reduc-
Xiong et al. 2002, Wang et al. 2017). Oxidative stress tase, glutathione S-transferase and peroxiredoxin (Miller
largely affects plant growth and development, crop qual- et al. 2010, Meyer et al. 2012, Noctor et al. 2014), and
ity and yield (Apel and Hirt 2004, Miller et al. 2010). non-enzymatic antioxidants including glutathione,
However, ROS are also necessary for inter- and intracel- ascorbic acid, carotenoids, tocopherols and flavonoids
lular signaling transduction and considered to be signal- are crucial for ROS homeostasis in plant (Gill and
ing molecules that regulate plant growth and Tuteja 2010). In this study, the enzyme activities of
development, adaptation to abiotic and biotic stresses POD, GPX and GR increased after severe alkali treatment
(Apel and Hirt 2004, Mittler et al. 2004). To keep (Fig. 4). Under mild alkali stress, no significant changes of

Physiol. Plant. 0, 2020 9

 alkali stresses in bermudagrass, indicating that amino acid
biosynthesis is a basic response of bermudagrass to severe
alkali stress. Mild alkali stress had little effect on the amino
acids, while severe alkali stress progressively increased the
content of amino acids (Fig. 6). These results indicated that
bermudagrass may invoke differential strategy in N metabo-
lism to counter the varying degrees of alkali stress.
Sugars have been characterized as important com-
pounds for the plant abiotic stress response (Klotke
et al. 2004). High carbohydrate accumulation under
stress condition is often indicative of good tolerance to
stress condition (Kerepesi and Galiba 2000). Soluble
sugars and sugar alcohols also function as compatible
solutes. They are responsible for regulating the osmotic
balance and protecting proteins and other macromole-
cules from damage caused by stress (Zhu 2002). Interest-
ingly, moderate and severe alkali stresses significantly
increased contents of disaccharide (sucrose, lactose, lac-
tulose, gentiobiose and maltose) and organic acid (buta-
noic acid, citric acid, maleic acid), but decreased
Fig. 8. A proposed model for alkali stress responses in bermudagrass. monosaccharides (fructose, allose, glucose, mannose
Mild alkali stress slowed down carbohydrate metabolism in and talose) and sugar alcohol (glycerol) (Figs 6 and 7).
bermudagrass, while alkali stress had the reversed effects. Nitrogen
These results indicated that carbon metabolism was par-
metabolisms were activated by both salt and alkali. Importantly, salt
and mild alkali-stressed plants accumulated significantly lower contents tially inhibited under severe alkali stress condition,
of ROS and MDA, had lower EL and higher survival rate than these which might lead to energy starvation and growth inhibi-
under moderated and severe alkali stress condition. Blue arrowhead: tion. Additionally, physiological analysis showed that
decrease; red arrowhead: increase; black arrowhead: no significant moderate or severe alkali treatment, but not mild alkali
changes.
treatment, significantly modulated osmolytes content,
ROS level and antioxidant enzyme activities (Figs 2-4).
Therefore, the bermudagrass may develop specific mech-
GSH content was observed, but moderate and severe anism to cope with mild alkali stress, such as restriction of
alkali stresses significantly increased GSH content carbohydrate consumption and N metabolism, to main-
(Fig. 3D). Physiological analysis also showed that both tain basic growth.
moderate and severe alkali treatments resulted in accu- Taken together, bermudagrass developed different
mulation of H2O2, O2•− and MDA contents (Fig. 3). approaches in response to mild, moderate and severe
These results indicated that bermudagrass under mild alkali stresses. Under mild alkali stress, bermudagrass
alkali stress condition suffered less oxidative stress com- might slow down metabolisms such as carbohydrate deg-
pared to severe alkali stress. Based on correlation analy- radation and N metabolism, resulting in the maintenance
sis, EL, MDA and H2O2 contents had strong correlation of basic growth but with a slower growth rate. However,
with other parameters which might be used as indicators moderate and severe alkali-stressed plant accumulated
of alkali stress tolerance in bermudagrass (Fig. 5). relative higher amount of carbohydrate which might
Inhibition/activation of specific metabolic pathways results in carbon starvation. Most importantly, mild alkali
under stress conditions may lead to stress-specific changes stresses slightly increased, while moderate and severe
in metabolite levels. Based on the results of metabolomic alkali conditions significantly increased ROS and MDA
data, we observed similar and contrasting changes of ber- contents (Fig. 8). Therefore, mild alkali-stressed plants
mudagrass in the response to mild and severe alkali stresses, exhibited relative lower EL, less growth inhibition and
implying the existence of differential metabolic rearrange- higher survival rate than these under moderate and
ments under different alkali stress conditions. Amino acids severe alkali stress conditions (Fig. 8).
are not only crucial for the synthesis of proteins, but they
can also act as precursors for a large number of metabolites
Author contributions
that have multiple functions in plant growth and response to
various stresses (Less and Galili 2008). Our results showed T.Y. and Z.C. designed the experiments. T.Y. and
that amino acid biosynthesis was commonly affected by Y.W. conducted the experiments. Y.Q.F. helped to

10 Physiol. Plant. 0, 2020

analyze and revise the manuscript. T.Y. and Z.C. wrote alkalinity stress in Kentucky Bluegrass (Poa pratensis).
the manuscript. All authors read and approved the Plant Mol Biol Report 33: 56–68
manuscript. Hu G, Liu Y, Duo T, Zhao B, Cui G, Ji J, Kuang X, Ervin EH,
Zhang X (2018) Antioxidant metabolism variation
associated with alkali-salt tolerance in thirty switchgrass
Acknowledgements – This research was supported by a
(Panicum virgatum) lines. PLoS One 13: e0199681
National Key Research and Development Program grant
Kerepesi I, Galiba G (2000) Osmotic and salt stress-induced
(2017YFD0201305) to Z.C. and Y.W., Huazhong Agricul-
alteration in soluble carbohydrate content in wheat
tural University Scientific & Technological Self-innovation
seedlings. Crop Sci 40: 482–487
Foundation (Program No.2016RC010) and the National
Klotke J, Kopka J, Gatzke N, Heyer AG (2004) Impact of
Natural Science Foundation of China (Grant
soluble sugar concentrations on the acquisition of freezing
No. 31872143) to Z.C.
tolerance in accessions of Arabidopsis thaliana with
contrasting cold adaptation-evidence for a role of
Data availability statement raffinose in cold acclimation. Plant Cell Environ 27:
1395–1404
The data that supports the findings of this study are avail- Kumari A, Parida AK (2018) Metabolomics and network
able in the supplementary material of this article analysis reveal the potential metabolites and biological
pathways involved in salinity tolerance of the halophyte
Salvadora persica. Environ Exp Bot 148: 85–99
References Less H, Galili G (2008) Principal transcriptional programs
Apel K, Hirt H (2004) Reactive oxygen species: metabolism, regulating plant amino acid metabolism in response to
oxidative stress, and signal transduction. Annu Rev Plant abiotic stresses. Plant Physiol 147: 316–330
Biol 55: 373–399 Li R, Shi F, Fukuda K, Yang Y (2010) Effects of salt and alkali
Bradford MM (1976) A rapid and sensitive method for the stresses on germination, growth, photosynthesis and ion
quantitation of microgram quantities of protein utilizing accumulation in alfalfa (Medicago sativa L.). Soil Sci Plant
the principle of protein-dye binding. Anal Biochem 72: Nutr 56: 725–733
248–254 Lisec J, Schauer N, Kopka J, Willmitzer L, Fernie AR (2015)
Chen M, Zhao Y, Zhuo C, Lu S, Guo Z (2015) Gas chromatography mass spectrometry-based
Overexpression of a NF-YC transcription factor from metabolite profiling in plants. Nat Protoc 10: 1457–1457
bermudagrass confers tolerance to drought and salinity in Lunde C, Drew DP, Jacobs AK, Tester M (2007) Exclusion of
transgenic rice. Plant Biotechnol J 13: 482–491 Na+ via sodium ATPase (PpENA1) ensures normal growth
Chen T, Johnson R, Chen S, Lv H, Zhou J, Li C (2018) of Physcomitrella patens under moderate salt stress. Plant
Infection by the fungal endophyte Epichloë bromicola Physiol 144: 1786–1796
enhances the tolerance of wild barley (Hordeum Meyer Y, Belin C, Delorme-Hinoux V, Reichheld JP,
brevisubulatum) to salt and alkali stresses. Plant Soil Riondet C (2012) Thioredoxin and glutaredoxin systems in
428: 1–18 plants, molecular mechanisms, crosstalks, and functional
Couée I, Sulmon C, Gouesbet G, El Amrani A (2006) significance. Antioxid Redox Signal 17: 1124–1160
Involvement of soluble sugars in reactive oxygen species Miller G, Suzuki N, Ciftci-Yilmaz S, Mittler R (2010) Reactive
balance and responses to oxidative stress in plants. J Exp oxygen species homeostasis and signalling during drought
Bot 57: 449–459 and salinity stresses. Plant Cell Environ 33: 453–467
Gill SS, Tuteja N (2010) Reactive oxygen species and Mittler R, Vanderauwera S, Gollery M, Van Breusegem F
antioxidant machinery in abiotic stress tolerance in crop (2004) Reactive oxygen gene network of plants. Trends
plants. Plant Physiol Biochem 48: 909–930 Plant Sci 9: 490–498
Guo R, Shi LX, Ding XM, Hu YJ, Tian SY, Yan DF, Shao SA, Noctor G, Mhamdi A, Foyer CH (2014) The roles of reactive
Gao YA, Liu R, Yang YF (2010) Effects of saline and oxygen metabolism in drought: not so cut and dried. Plant
alkaline stress on germination, seedling growth, and ion Physiol 164: 1636–1648
balance in wheat. Agron J 102: 1252–1260 Obata T, Fernie AR (2012) The use of metabolomics to
Guo R, Yang Z, Li F, Yan C, Zhong X, Liu Q, Xia X, Li H, dissect plant responses to abiotic stresses. Cell Mol Life Sci
Zhang L (2015) Comparative metabolic responses and 69: 3225–3243
adaptive strategies of wheat (Triticum aestivum) to salt and Pastori GM, Foyer CH (2002) Common components,
alkali stress. BMC Plant Biol 15: 170 networks, and pathways of cross-tolerance to stress. The
de Hoon MJL, Imoto S, Nolan J, Miyano S (2004) Open central role of "redox" and abscisic acid-mediated
source clustering software. Bioinformatics 20: 1453–1454 controls. Plant Physiol 129: 460–468
Hu LX, Zhang PP, Jiang Y, Fu JM (2015) Metabolomic Shi H, Wang Y, Cheng Z, Ye T, Chan Z (2012) Analysis of
analysis revealed differential adaptation to salinity and natural variation in bermudagrass (Cynodon dactylon)

Physiol. Plant. 0, 2020 11

 reveals physiological responses underlying drought Xiong L, Schumaker KS, Zhu JK (2002) Cell signaling during
tolerance. PLoS One 7: e53422 cold, drought, and salt stress. Plant Cell 14: S165–S183
Shi H, Ye T, Chan Z (2013) Comparative proteomic and Yang CW, Xu HH, Wang LL, Liu J, Shi DC, Wang DL (2009)
physiological analyses reveal the protective effect of Comparative effects of salt-stress and alkali-stress on the
exogenous polyamines in the bermudagrass (Cynodon growth, photosynthesis, solute accumulation, and ion
dactylon) response to salt and drought stresses. J Proteome balance of barley plants. Photosynthetica 47: 79–86
Res 12: 4951–4964 Ye T, Shi H, Wang Y, Chan Z (2015) Contrasting changes
Shi H, Jiang C, Ye T, Tan DX, Reiter RJ, Zhang H, Liu R, caused by drought and submergence stresses in
Chan Z (2015) Comparative physiological, metabolomic, bermudagrass (Cynodon dactylon). Front Plant Sci
and transcriptomic analyses reveal mechanisms of 6: 951
improved abiotic stress resistance in bermudagrass Ye T, Shi H, Wang Y, Yang F, Chan Z (2016) Contrasting
[Cynodon dactylon (L). Pers.] by exogenous melatonin. proteomic and metabolomic responses of
J Exp Bot 66: 681–694 bermudagrass to drought and salt stresses. Front Plant
Teng K, Tan P, Guo W, Yue Y, Fan X, Wu J (2018) Sci 7: 1694
Heterologous expression of a novel Zoysia japonica C2H2 Zhang P, Fu JM, Hu LX (2012) Effects of alkali stress on
zinc finger gene, ZjZFN1, improved salt tolerance in growth, free amino acids and carbohydrates metabolism
Arabidopsis. Front Plant Sci 9: 1159 in Kentucky bluegrass (Poa pratensis). Ecotoxicology 21:
Theerawitaya C, Yamada-Kato N, Singh HP, Cha-um S, 1911–1918
Takabe T (2018) Isolation, expression, and functional Zhu JK (2002) Salt and drought stress signal transduction in
analysis of developmentally regulated plasma membrane plants. Annu Rev Plant Biol 53: 247–273
polypeptide 1 (DREPP1) in Sporobolus virginicus grown
under alkali salt stress. Protoplasma 255: 1–10 Supporting information
Wang GY, Bi AY, Amombo E, Li HY, Zhang L, Chen C, Hu T,
Additional supporting information may be found online
Fu JM (2017) Exogenous calcium enhances the
in the Supporting Information section at the end of the
photosystem II photochemistry response in salt stressed
article.
tall fescue. Front Plant Sci 8: 2032
Wu DZ, Shen QF, Cai SG, Chen ZH, Dai F, Zhang GP (2013) Table S1. Preparation of salt and alkali solutions.
Ionomic responses and correlations between elements
and metabolites under salt stress in wild and cultivated Table S2. Metabolite contents and fold changes after
barley. Plant Cell Physiol 54: 1976–1988 alkali treatments.

Edited by B. Huang

12 Physiol. Plant. 0, 2020