免费文献传递   相关文献

Plants Adapt to LongTerm Potassium Deficiency by Accumulation of Membrane Lipids in Leaves and Maintenance of Lipid Composition in Roots

植物适应长期缺钾——积累叶片膜脂、维持根膜脂组成不变



全 文 :植物适应长期缺钾
———积累叶片膜脂、 维持根膜脂组成不变∗
王丹丹1ꎬ2ꎬ 郑国伟1ꎬ 李唯奇1ꎬ3∗∗
(1 中国科学院昆明植物研究所中国西南野生生物种质资源库ꎬ 云南 昆明  650201ꎻ 2 中国科学院大学ꎬ
北京  100049ꎻ 3 红河学院生物系ꎬ 云南 蒙自  661100)
摘要: 环境对植物的胁迫可能是短期快速的、 也可能是长期而缓慢的ꎬ 而植物应对这两种胁迫的策略可能
不同ꎮ 膜脂组成变化是植物响应环境胁迫的主要手段之一ꎬ 其响应长期胁迫和短期胁迫的样式也可能不
同ꎮ 植物膜脂组成对短期缺钾胁迫的响应已经有报道ꎬ 但是对长期缺钾的响应如何尚且未知ꎮ 我们设置了
4种 (5􀆰 1ꎬ 0􀆰 51ꎬ 0􀆰 051和 0 mmol􀅰L-1) 不同的钾浓度ꎬ 比较了拟南芥 (Arabidopsis thaliana) 及其生长于
贫钾生境中的近缘种须弥芥 (Crucihimalaya himalaica) 长期缺钾后 (18 天) 的生理和生化变化ꎬ 发现须
弥芥具有耐受贫钾的能力ꎮ 我们进一步运用脂类组学的方法检测比较了拟南芥和须弥芥在长期缺钾胁迫下
脂类组成的变化ꎬ 发现: (1) 两种植物叶片中总脂以及几乎所有脂类的含量明显上升ꎻ (2) 两种植物都是
地上部分膜脂的变化幅度大于根部膜脂的变化幅度ꎻ (3) 地上部分膜脂变化幅度ꎬ 须弥芥的大于拟南芥
的ꎻ 地下部分的膜脂变化幅度ꎬ 须弥芥的小于拟南芥的ꎻ (4) 拟南芥叶片和根中 PA的含量显著上升ꎬ 与
PA相对应的是 PE含量的显著下降ꎬ 由此我们推测拟南芥中 PA 的积累主要来自于 PE 的水解ꎮ 上述结果
提示ꎬ 在细胞水平上ꎬ 植物主要通过积累叶片膜脂和维持根部膜脂组成基本不变来适应长期缺钾ꎮ
关键词: 须弥芥ꎻ 拟南芥ꎻ 耐缺钾ꎻ 膜脂
中图分类号: Q 945            文献标识码: A              文章编号: 2095-0845(2014)02-163-14
Plants Adapt to Long ̄Term Potassium Deficiency by Accumulation
of Membrane Lipids in Leaves and Maintenance of
Lipid Composition in Roots
WANG Dan ̄Dan1ꎬ2ꎬ ZHENG Guo ̄Wei1ꎬ LI Wei ̄Qi1ꎬ3∗∗
(1 Germplasm Bank of Wild Species in Southwest Chinaꎬ Kunming Institute of Botanyꎬ Chinese Academy of Sciencesꎬ
Kunming 650201ꎬ Chinaꎻ 2 University of Chinese Academy of Sciencesꎬ Beijing 100049ꎬ Chinaꎻ
3 Biology Departmentꎬ Honghe Universityꎬ Mengzi 661100ꎬ China)
Abstract: Environmental stresses on plants can be divided into short ̄ and long ̄term typesꎬ which may be associated
with different adaptation strategies. Adjustment of the composition of membrane lipid is a major response to stress.
The membrane lipid composition may different between short ̄ and long ̄term environment stresses. A previous study
reported changes in the lipid composition in barley root under short ̄term potassium (K+) deficiencyꎻ howeverꎬ the
equivalent response of plants to long ̄term K+ deficiency remains completely unknown. Plants of Arabidopsis thaliana
and Crucihimalaya himalaica (Brassicaceae) were grown at four different K+ levels ( 5􀆰 1ꎬ 0􀆰 51ꎬ 0􀆰 051 and 0
mmol􀅰L-1) for 18 days. Physiological and biochemical experiments were conducted on this issue and the results sug ̄
gest that C􀆰 himalaicaꎬ a relative of A􀆰 thalianaꎬ derived from a K+  ̄deficient areaꎬ is tolerant to K+  ̄limited condi ̄
植 物 分 类 与 资 源 学 报  2014ꎬ 36 (2): 163~176
Plant Diversity and Resources                                    DOI: 10.7677 / ynzwyj201413099

∗∗
Funding: The National Natural Science Foundation of China (NSFC 30670474ꎬ 30870571 and 31070262)
Author for correspondenceꎻ E ̄mail: weiqili@mail􀆰 kib􀆰 ac􀆰 cn
Received date: 2013-04-25ꎬ Accepted date: 2013-07-18
作者简介: 王丹丹 (1983-) 女ꎬ 助理研究员ꎬ 主要从事植物逆境分子生理学研究ꎮ E ̄mail: wangdandan@mail􀆰 kib􀆰 ac􀆰 cn
tions. Electrospray ionization tandem mass spectrometry (ESI ̄MS / MS) was used to determine the lipid changes in
A􀆰 thaliana and C􀆰 himalaica subjected to long ̄term K+ deficiency. The results showed that: (1) the levels of total
lipids and most lipid classes in leaves of A􀆰 thaliana and C􀆰 himalaica increased under K+  ̄deficient conditionsꎻ (2)
the changes in lipid content in leaves of A􀆰 thaliana and C􀆰 himalaica were greater than those in the rootsꎻ (3) the
change in lipid content in leaves of C􀆰 himalaica was greater than that in A􀆰 thalianaꎬ with the opposite trend being
shown in the roots and (4) in A􀆰 thalianaꎬ the increase in phosphatidic acid (PA) corresponded to the decrease in
phosphatidylethanolamine (PE). This indicates that K+  ̄deficiency ̄induced PA in A􀆰 thaliana was derived primarily
from PE. Our results suggest thatꎬ at the cellular levelꎬ plants adapt to long ̄term K+ deficiency by the accumulation
of lipids in leaves and maintenance of the lipid composition in roots.
Key words: Crucihimalaya himalaicaꎻ Arabidopsis thalianaꎻ Potassium deficiency toleranceꎻ Membrane lipids
Abbreviations: digalactosyldiacylglycerolꎬ DGDGꎻ monogalactosyldiacylglycerolꎬ MGDGꎻ phosphatidylglycerolꎬ
PGꎻ phosphatidylcholineꎬ PCꎻ phosphatidylethanolamineꎬ PEꎻ phosphatidylinositolꎬ PIꎻ phosphatidylserineꎬ PSꎻ
phosphatidic acidꎬ PAꎻ lysophospholipidꎬ lysoPLꎻ lysophosphatidylglycerolꎬ lysoPGꎻ lysophosphatidylcholineꎬ ly ̄
soPCꎻ lysophosphatidylethanolamineꎬ lysoPEꎻ electrospray ionization tandem mass spectrometryꎬ ESI ̄MS / MSꎻ
phospholipidꎬ PLꎻ glycolipidꎬ GLꎻ leafꎬ Lꎻ rootꎬ R
  As sessile organismsꎬ plants must cope with va ̄
rious environmental stressesꎬ such as droughtꎬ salin ̄
ityꎬ nutrient deficiency and extreme temperatures.
Nitrogen (N)ꎬ phosphorus and potassium (K+) are
the essential macronutrients of plants. Low availabili ̄
ty of these three elements is a major constraint for
crop production in many agriculture systems world ̄
wide (Chen et al.ꎬ 2008). In developing countriesꎬ
the importance of K+ has sometimes been overlookedꎬ
and financial constraints have forced farmers to pri ̄
oritize the application of nitrogen over K+ . As a re ̄
sultꎬ a considerable area of farmland has become
K+  ̄deficient (Armengaud et al.ꎬ 2009). In Chinaꎬ
about one ̄quarter to one ̄third of arable soils and
three ̄quarters of paddy soils are considered K+  ̄defi ̄
cient (Luꎬ 1989ꎻ Yang et al.ꎬ 2004).
There are various types of environmental stressꎬ
and the same stress can influence plants in different
ways. For exampleꎬ a single form of stress can be di ̄
vided into short ̄ and long ̄term types based on its
temporal scale. Taking temperature stress as an ex ̄
ampleꎬ perennial plants are subjected to long ̄term
cold stress in winter and long ̄term heat stress in
summer. For plants growing in alpine and desert eco ̄
systemsꎬ there are short periods of high temperature
in the daytime and low temperature at night during
the summer. Accordinglyꎬ plants have evolved so ̄
phisticated mechanisms to adapt to short ̄ and long ̄
term environmental stressesꎬ not only at the levels of
phenologyꎬ morphology and physiologyꎬ but also at
the biochemical level.
Membranesꎬ particularly plasma and chloroplast
membranes are sensitive to environmental stimuli.
Glycerolipids are the major constituents of cellular
membranes. Plants adapt to environmental stress by
adjusting the compositionꎬ double bond index and
acyl chain length ( ACL) of glycerolipids. It has
been reported that phosphatidic acid ( PA) has a
tendency to form a hexagonal II phase (Verkleij et
al.ꎬ 1982)ꎬ which has a strong propensity to result
in membrane leakage. Many studies have suggested
that PA is negatively related to membrane integrity
(Li et al.ꎬ 2008ꎻ Welti et al.ꎬ 2002ꎻ Zhang et al.ꎬ
2012). It is well known that the degree of glycerolip ̄
id unsaturation affects membrane fluidity. Increase or
decrease in the degree of unsaturation of glycerolip ̄
ids was shown to be associated with enhancement or
reduction of the fluidity of membranes as an adapta ̄
tion to long ̄term cold stress and heat stressꎬ respec ̄
tively. In contrastꎬ plants adapt to short ̄term and
rapid temperature fluctuation by lipid remodelling
rather than alteration of the degree of unsaturation
(Zheng et al.ꎬ 2011).
As for nutrient stressꎬ short ̄term phosphate dep ̄
461                                  植 物 分 类 与 资 源 学 报                            第 36卷
rivation (for 4 h) resulted in an increase in PA levelꎬ
but after long ̄term phosphate deprivation stress ( for
18 days)ꎬ a decrease in PA level was observed
(Russo et al.ꎬ 2007). This is because PA has dual
roles in the response of plants to stress. As a second ̄
ary messengerꎬ increases in PA levels have been ob ̄
served under various stress conditionsꎬ such as
drought ( Hong et al.ꎬ 2010 )ꎬ low ̄temperature
stress (Zhang et al.ꎬ 2012) and nitrogen signalling
(Hong et al.ꎬ 2009). Howeverꎬ as a phosphorus ̄
containing lipidꎬ a reduction in the level of PA is an
inevitable result of long ̄term phosphate deprivation.
As for potassium deficiency stressꎬ phospholipid
(PL) did not change with short ̄term K+ deprivation
treatment (24 hꎬ 30 h and 36 h). With respect to in ̄
dividual types of PLꎬ 30 h of K+ deprivation led to
reductions in phosphatidylcholine (PC)ꎬ phosphati ̄
dylserine ( PS ) and phosphatidylinositol ( PI )ꎬ
whereas phosphatidylglycerol (PG)ꎬ phosphatidyle ̄
thanolamine (PE) and PA levels increased (Hafsi
et al.ꎬ 2009). When there is a deficiency of K+ꎬ
plant growth is retardedꎬ and net retranslocation of
K+ from mature leaves and stems is enhancedꎻ under
severe deficiencyꎬ these organs become chlorotic and
necrotic (Marschnerꎬ 1995). Alternative cations such
as Na+ and Mg2+ are accumulated in the vacuole as
replacements for K+ to maintain turgor. When these
cations are absentꎬ organic solutes must be accumula ̄
ted (Leigh and Wyn Jonesꎬ 1984). In additionꎬ it
has been reported that potassium starvation activates
K+ uptake in plant roots by regulating the activity of
potassium transport proteins (Ashley et al.ꎬ 2006).
Microarray and deep sequencing technologies
have been employed to reveal the molecular mecha ̄
nisms that plants use to adapt to K+ deficiency stress
(Armengaud et al.ꎬ 2004ꎻ Gierth et al.ꎬ 2005ꎻ Ma
et al.ꎬ 2012ꎻ Wang et al.ꎬ 2012). These studies
showed that fewer genes exhibited transcriptional
changes in response to K+ deficiency than to phos ̄
phorus or nitrogen deficiency (Gierth et al.ꎬ 2005ꎻ
Ma et al.ꎬ 2012). It has also been reported that
there are highly significant K+  ̄responsive genes in
Arabidopsis that encode cell wall proteins and ion
transportersꎬ as well as proteins with a putative role
in Ca2+ signallingꎻ in additionꎬ jasmonic acid ̄related
genes showed the largest response to K+ deficiency
(Armengaud et al.ꎬ 2004). Comparative analysis of
transcriptomic differences between Arabidopsis and
rice showed that monocots and dicots share similar
mechanisms in response to K+ deficiency (Ma et al.ꎬ
2012). Transcriptome sequencing of two soybean
genotypes also provided preliminary information on
the molecular mechanism of potassium absorption
and transport under conditions of low ̄K+ stress (Wang
et al.ꎬ 2012 ). Although certain aspects of lipid
changes in plants subjected to short ̄term K+ defi ̄
ciency are knownꎬ the way in which membrane lipid
composition responds to long ̄term K+ deficiency re ̄
mains unclear.
Alpine screes are special ecosystem of alpine
areasꎬ which including several extreme environ ̄
mentsꎬ such as strong solar radiation and low oxy ̄
gen. Thin and sterile soil layer is another extreme
environment of alpine screes. The depth of the A ̄ho ̄
rizon in alpine screes in Baima Snow Mountain is
just 0-5 cm and the available potassium is 99 mg / kg
(Li et al.ꎬ 2003)ꎬ which is relatively low (Guanꎬ
2001). Plants grow in this environment type may
have the ability to tolerant to potassium deficiency.
C􀆰 himalaicaꎬ a relative of A􀆰 thaliana (Al ̄Shehbaz
et al.ꎬ 2006)ꎬ with which it shares morphological
similaritiesꎬ is derived from the alpine screes in Bai ̄
ma Snow Mountain and may be tolerant to K+ defi ̄
ciency. In view of these characteristicsꎬ a compara ̄
tive study of the responses of C􀆰 himalaica and
A􀆰 thaliana to K+ deficiency could provide a deeper
insight into the adaptive mechanisms of plants that
lead to tolerance to K+ deficiency.
Plant lipidomics is based on electrospray ioniza ̄
tion tandem mass spectrometry (ESI ̄MS / MS) analy ̄
sis. It requires only simple preparation and small
samples to identify and quantify 11 head ̄group lipid
classes and more than 140 lipid molecular speciesꎬ
thereby providing a rapid means of determining the
5612期    WANG Dan ̄Dan et al.: Plants Adapt to Long ̄Term Potassium Deficiency by Accumulation of Membrane 􀆺    
lipid composition of different plant species under va ̄
rious growth conditions. Lipidomics has been applied
to profile changes in lipid composition in plants sub ̄
jected to temperature stress (Chen et al.ꎬ 2006ꎻ Li
et al.ꎬ 2004ꎻ Welti et al.ꎬ 2002)ꎬ nutrient stress
(Li et al.ꎬ 2006aꎬ b)ꎬ water stress (Hong et al.ꎬ
2008) and heavy metal toxicity ( Zhao et al.ꎬ
2011). In the present studyꎬ physiological experi ̄
ments were conducted on both A􀆰 thaliana and
C􀆰 himalaica under various K+ levelsꎬ as well as
comparative analysis of the profiles of membrane lip ̄
ids between these two species. From these analysesꎬ
we confirmed that C􀆰 himalaica has a greater toler ̄
ance to K+ deficiency than A􀆰 thaliana. We also found
that the responses of lipid compositions in roots to
K+ deficiency differ from those in leaves. The pat ̄
terns of lipid change were similar in A􀆰 thaliana and
C􀆰 himalaicaꎬ but the extents of the changes differed.
1  Methods and materials
1􀆰 1  Plant materials and growth conditions
Arabidopsis thaliana ( Columbia ecotype) and
Crucihimalaya himalaica were used for all experi ̄
ments. Seeds were surface ̄sterilized with 70% (v / v)
ethanol and 5% NaClOꎬ followed by washing with
double ̄distilled water three times. The sterilized
seeds were cold stratificated at 4 ℃ in the dark for
two and five days for A􀆰 thaliana and C􀆰 himalaicaꎬ
respectively. Seeds were then planted on Murashige
and Skoog medium and transferred to a growth cham ̄
ber at 22 ℃ with 12 h daylight at 120 μmol m-2s-1 .
After 15 days of growthꎬ plantlets were transferred to
a hydroponic system (Tocquin et al.ꎬ 2003) for fur ̄
ther growth. Plants of A􀆰 thaliana and C􀆰 himalaica
with the same rosette diameter (30 days old and 35
days old for A􀆰 thaliana and C􀆰 himalaicaꎬ respec ̄
tively) were used throughout the study. Plants were
transferred to a modified hydroponic system with dif ̄
ferent potassium levels: K+ sufficiency (5􀆰 1 mmol􀅰
L-1 KNO3 )ꎬ mild K
+ limitation ( 0􀆰 51 mmol􀅰L-1
KNO3)ꎬ severe K
+ limitation ( 0􀆰 051 mmol􀅰L-1
KNO3) and K
+ deprivation (0 mmol􀅰L-1 KNO3).
The nitrogen level was balanced with NH4NO3 . The
growth medium was changed every six days. On day
18ꎬ shoots and roots were harvested seperately and
stored at -80 ℃ or used immediately for analysis.
1􀆰 2  Phenotype assay
For each plant speciesꎬ five replicates were used
to generate meansꎻ each replicate consisted of a pool
of one to three plants. Photographs were taken on
day 18. Shoots and roots were harvested separately
and dried at 80 ℃ until they reached a constant
weight. The dry shoots and roots were then weighed.
The number of leaves was also counted and recor ̄
ded. Rosette diameter was measured using a rulerꎬ
and each plant was measured three times at different
orientations to generate the means.
1􀆰 3  Detection of total potassium
Shoots and roots were harvested and dried sepa ̄
ratelyꎬ and then ground to a fine powder with a pes ̄
tle. The plant powder was digested with HNO3 ̄
HClO4 and then dissolved in HCl. Potassium content
was determined using an inductively coupled plasma
atomic ̄emission spectrometer ( IRIS Advantage ̄ERꎻ
Thermo Jarrell Ash Corporationꎬ USA).
1􀆰 4  Lipid extraction and ESI ̄MS / MS analysis
Leaves and roots from the seedlings were collect ̄
ed and immediately transferred into isopropanol with
0􀆰 1% butylated hydroxytoluene at 75 ℃ to inhibit
lipolytic activity. Lipid extractionꎬ ESI ̄MS / MS anal ̄
ysis and quantification were performed as previously
described (Welti et al.ꎬ 2002). The Q ̄test was per ̄
formed on the total amount of lipid in each classꎬ and
data from discordant samples were removed. The data
were subjected to one ̄way factorial ANOVA with
SPSS 16􀆰 0. Differences between means were tested by
Fisher’s least significant difference (LSD) method.
2  Results and discussion
2􀆰 1  C􀆰 himalaica exhibited higher tolerance to
low potassium supply than A􀆰 thaliana
Seedlings of A􀆰 thaliana and C􀆰 himalaica were
used throughout this study. Seeds of C􀆰 himalaica were
harvested from Baima Snow Mountain. A􀆰 thaliana and
661                                  植 物 分 类 与 资 源 学 报                            第 36卷
C􀆰 himalaica plants were precultured in a hydroponic
system as described previously ( Tocquin et al.ꎬ
2003). Thenꎬ 30 ̄ and 35 ̄day ̄old seedlings of these
two species were transplanted to hydroponic media
with different levels of K+ . The leaves and roots of the
species were harvested separately after 18 days of
growth. Figure 1 shows the growth of A􀆰 thaliana and
C􀆰 himalaica for four levels of potassium: 5􀆰 1 (con ̄
trol)ꎬ 0􀆰 51 (mild limitation)ꎬ 0􀆰 051 (severe limita ̄
tion) and 0 mmol􀅰L-1 KNO3 (deprivation). Under
K+ ̄deficient conditionsꎬ plants of both species
showed retarded growth and a chlorosis phenotypeꎬ
but C􀆰 himalaica exhibited less chlorosis than
A􀆰 thaliana under 0􀆰 051 and 0 mmol􀅰L-1 potassium
conditions (Fig􀆰 1). C􀆰 himalaica also showed less re ̄
duction in rosette diameter (Fig􀆰 2A)ꎬ leaf number
(Fig􀆰 2B) and total plant dry weight (DW) (Fig􀆰 2C)
under K+  ̄limited conditions than Arabidopsis. These
results indicate that C􀆰 himalaica grew better than
A􀆰 thaliana under K+  ̄limited conditions.
In contrast to nitrogen and phosphorus deficien ̄
cyꎬ plants rarely increase their root biomass under
K+  ̄deficient conditions. This is likely to be a conse ̄
quence of impaired sucrose export from the leaves of
K+  ̄deficit plants (Cakmak et al.ꎬ 1994ꎻ Hermans et
al.ꎬ 2006). In this studyꎬ the root / shoot (R ∶ S) dry
weight ratios in A􀆰 thaliana were unchanged under
K+  ̄deficient conditions. In contrastꎬ these ratios in
C􀆰 himalaica significantly increasedꎬ with 25􀆰 08%ꎬ
53􀆰 09% and 55􀆰 62% increases under 0􀆰 51ꎬ 0􀆰 051
and 0 mmol􀅰L-1 K+ conditionsꎬ respectivelyꎬ com ̄
pared with the control (Fig􀆰 2D). Generallyꎬ greater
root growth and modified root morphology enables
plants to obtain minerals that are present at low levels
in the rhizosphere more effectively (Hermans et al.ꎬ
2006). Thusꎬ the increased R ∶ S dry weight ratios in
C􀆰 himalaica might be one of the reasons that it can
maintain better growth under K+ ̄deficient conditions.
Another possible reason for the better growth of
C􀆰 himalaica under K+  ̄deficient conditions is that it
has a higher capacity to acquire K+ . To test this pos ̄
sibilityꎬ we analysed the total K+ content in shoots
and roots. K+ deficiency led to a marked decrease in
the potassium content in leaves and roots of both
species. Shoots of C􀆰 himalaica had higher total K+
content than A􀆰 thaliana under both control and K+  ̄
limited conditions (Fig􀆰 3A). A􀆰 thaliana showed a
greater decrease in potassium content than C􀆰 himalaica.
In Arabidopsis shootsꎬ K+ content was reduced by
53􀆰 13%ꎬ 67􀆰 19% and 68􀆰 37% at 0􀆰 51ꎬ 0􀆰 051 and
0 mmol􀅰L-1 K+ treatmentsꎬ respectivelyꎬ compared
with the control. Howeverꎬ decreases of only
32􀆰 46%ꎬ 48􀆰 21% and 60􀆰 96% occurred in shoots of
C􀆰 himalaica (Fig􀆰 3A).
Fig􀆰 1  Phenotype comparison of A􀆰 thaliana (AT) and C􀆰 himalaica (CH). The plants were continuously grown at four different K+ levels
(sufficient K+: 5􀆰 1 mmol􀅰L-1ꎬ mildly limited K+: 0􀆰 51 mmol􀅰L-1ꎬ severely limited K+: 0􀆰 051 mmol􀅰L-1
and K+ deprivation: 0 mmol􀅰L-1) for 18 days
7612期    WANG Dan ̄Dan et al.: Plants Adapt to Long ̄Term Potassium Deficiency by Accumulation of Membrane 􀆺    
Fig􀆰 2  Effects of different potassium levels on the growth of A􀆰 thaliana and C􀆰 himalaica. Data are means ± SD (n= 4 or 5) . Bars with different
letters differ significantly at P < 0􀆰 05 (Fisher’s least significant difference) . Black barsꎬ A􀆰 thalianaꎻ light grey barsꎬ C􀆰 himalaica
A. Changes of rosette diameter under different K+ conditionsꎻ B. Changes of leaf number under different K+ conditionsꎻ C. Changes
of total plant dry weight under different K+ conditionsꎻ D. Changes of root: shoot ratio under different K+ conditions
Fig􀆰 3  Changes in potassium content of A􀆰 thaliana and C􀆰 himalaica under different K+ conditions. Data are means ± SD (n= 4 or 5) . Bars with
different letters differ significantly at P < 0􀆰 05 (Fisher’s least significant difference) . Black barsꎬ A􀆰 thalianaꎻ light grey barsꎬ C􀆰 himalaica
A. Changes of total potassium content under different K+ conditionsꎻ B. Changes of shoot potassium
to root potassium ratio under different K+ conditions
861                                  植 物 分 类 与 资 源 学 报                            第 36卷
    At the root levelꎬ potassium content decreased
with decreasing potassium supply (Fig􀆰 3A). C􀆰 him ̄
alaica absorbed more potassium than A􀆰 thaliana un ̄
der both K+ ̄sufficient and K+ ̄deficient conditions. In
contrast to the shootsꎬ the decreases of total K+ con ̄
tent of C􀆰 himalaica were greater than those of
A􀆰 thaliana under K+ ̄limited conditions. Under 0􀆰 51ꎬ
0􀆰 051 and 0 mmol􀅰L-1 K+ conditionsꎬ C􀆰 himalaica
exhibited 43􀆰 23%ꎬ 68􀆰 16% and 71􀆰 96% reductions
in total K+ contentꎬ respectivelyꎬ compared with the
controlꎬ while the reductions of A􀆰 thaliana were
41􀆰 09%ꎬ 57􀆰 61% and 62􀆰 09%. We further calculat ̄
ed the ratios of shoot K+ / root K+ of both species. K+
deficiency resulted in a significant decrease in shoot
K+ / root K+ ratios in A􀆰 thaliana (Fig􀆰 3B). In con ̄
trastꎬ these ratios of C􀆰 himalaica increased dramatic ̄
ally under K+ ̄limited conditions. This indicated thatꎬ
in C􀆰 himalaicaꎬ under K+ ̄limited conditionsꎬ a con ̄
siderable proportion of K+ was allocated to the shoots.
This may be one of the reasons for less chlorosis in
C􀆰 himalaica leaves under K+ ̄limited conditions.
2􀆰 2  Profiling and quantification of membrane
lipids in A􀆰 thaliana and C􀆰 himalaica plants un ̄
der different potassium conditions
To obtain a comprehensive understanding of how
the lipid composition of plant membranes responds to
long ̄term K+ deficiencyꎬ we used a lipidomics ap ̄
proach based on ESI ̄MS / MS to profile and quantify
the lipids in A􀆰 thaliana and C􀆰 himalaica. Two ga ̄
lactolipids: digalactosyldiacylglycerol (DGDG) and
monogalactosyldiacylglycerol (MGDG)ꎬ and six phos ̄
pholipids: phosphatidylglycerol ( PG)ꎬ phosphati ̄
dylcholine (PC)ꎬ phosphatidylethanolamine (PE)ꎬ
phosphatidylinositol (PI)ꎬ phosphatidylserine (PS)
and phosphatidic acid ( PA)ꎬ categories which in ̄
clude 140 molecular species of membrane glycerolip ̄
idsꎬ were detected. DGDGꎬ MGDG and PG are the
major constituents of plastidic membrane lipidsꎬ
whereas PCꎬ PEꎬ PIꎬ PS and PA are extraplastidic
lipids (Li et al.ꎬ 2008).
Absolute amounts (nmol􀅰mg-1 dry weight) of
these lipidsꎬ which reflect lipid degradation or accu ̄
mulationꎬ are shown in Tables 1-3 and Fig􀆰 4. Rela ̄
tive amounts (mol%) of total glycolipids (GLs) and
phospholipids (PLs)ꎬ which can reflect the overall
composition of membrane lipidꎬ are shown in Table 4.
The total amounts of lipid in each head ̄group class
and total lipids in leaves and roots of both species are
shown in Tables 1 and 2ꎬ respectively. The total a ̄
mounts of GL and PL in the leaves and roots of both
species are shown in Table 3. In additionꎬ the a ̄
mounts of each molecular species in the leaves and
roots of both species are displayed in Fig􀆰 4.
Overallꎬ we found thatꎬ under K+  ̄sufficient
conditionsꎬ the amounts of lipids in each head group
class and total lipids in A􀆰 thaliana leaves and roots
were higher than those in C􀆰 himalaica. Under K+  ̄
limited conditionsꎬ the amounts of membrane lipids
increased in A􀆰 thaliana and C􀆰 himalaica leaves
(Tables 1 and 3)ꎬ but no evident change in lipids
was observed in their roots (Tables 2 and 3). Most
lipid molecular species in the leaves of A􀆰 thaliana
and C􀆰 himalaica exhibited an increase after long ̄
term K+ deficiency (Fig􀆰 4A and 4B)ꎬ whereas the
changes of lipid molecular species in the roots of
these species were variable (Fig􀆰 4C and 4D). De ̄
tailed findings on the changes in lipids under K+  ̄de ̄
ficient conditions are described below.
2􀆰 3  The amounts of almost all membrane lipids
greatly increased in A􀆰 thaliana leaves after 18 ̄
day K+ ̄deficient treatment
Compared with plants grown under control con ̄
ditionsꎬ plants of A􀆰 thaliana grown under K+  ̄limited
conditions showed marked increases in the amount of
lipids (Table 1). The total amount showed 1􀆰 44%ꎬ
35􀆰 71% and 30􀆰 76% increases under 0􀆰 51ꎬ 0􀆰 051
and 0 mmol􀅰L-1 K+ conditionsꎬ respectivelyꎬ com ̄
pared with the control. The plastidic lipids DGDGꎬ
MGDG and PG showed greater increases than other
types. This was most pronounced for MGDGꎬ with
2%ꎬ 41􀆰 64% and 38􀆰 21% increases under 0􀆰 51ꎬ
0􀆰 051 and 0mmol􀅰L-1 K+ conditionsꎬ respectively. We
also found that the amounts of total GLs and total PLs
in K+  ̄deficient A􀆰 thaliana leaves were significantly
9612期    WANG Dan ̄Dan et al.: Plants Adapt to Long ̄Term Potassium Deficiency by Accumulation of Membrane 􀆺    
Table 1  Total amount of lipid in each head ̄group class and total lipids in A􀆰 thaliana and C􀆰 himalaica leaves under different K+ levels.
Data are means ± SD (n= 4 or 5) . Values in the same row with different letters differ significantly at P < 0􀆰 05 (Fisher’s least significant
difference) . Asterisk indicates that the value is different from that of A􀆰 thaliana under the same conditions (P < 0􀆰 05)
Lipid
class
  Plant
  species
Lipid / nmol􀅰mg-1 DW
Control     
0􀆰 51 
mmol􀅰L-1  
0􀆰 051
mmol􀅰L-1
  0
  mmol􀅰L-1
Relative change / %
0􀆰 51
mmol􀅰L-1
0􀆰 051
mmol􀅰L-1

mmol􀅰L-1
DGDG A􀆰 thaliana 47􀆰 9±2􀆰 07b 47􀆰 6±10􀆰 5b 60􀆰 9±3􀆰 30a 59􀆰 4±1􀆰 04a -0􀆰 63 27􀆰 14 24􀆰 01
C􀆰 himalaica 41􀆰 3±0􀆰 77b 56􀆰 1±3􀆰 11a∗ 59􀆰 7±5􀆰 66a 54􀆰 8±5􀆰 69a 35􀆰 51 44􀆰 20 32􀆰 37
MGDG A􀆰 thaliana 195􀆰 0±15􀆰 78b 198􀆰 9±54􀆰 29b 276􀆰 2±25􀆰 95a 269􀆰 5±24􀆰 39a 2􀆰 00 41􀆰 64 38􀆰 21
C􀆰 himalaica 144􀆰 2±9􀆰 67b∗ 218􀆰 0±8􀆰 08a 223􀆰 0±26􀆰 96a∗ 205􀆰 3±16􀆰 28a∗ 51􀆰 18 54􀆰 65 42􀆰 37
PG A􀆰 thaliana 18􀆰 6±0􀆰 41b 21􀆰 5±1􀆰 66ab 23􀆰 2±2􀆰 35a 19􀆰 9±5􀆰 13b 15􀆰 59 24􀆰 73 6􀆰 99
C􀆰 himalaica 8􀆰 72±0􀆰 59b∗ 16􀆰 2±1􀆰 23a∗ 14􀆰 8±2􀆰 03a∗ 14􀆰 02±1􀆰 22a∗ 85􀆰 78 69􀆰 72 60􀆰 55
PC A􀆰 thaliana 20􀆰 8±1􀆰 69c 27􀆰 6±1􀆰 06ab 32􀆰 5±1􀆰 29a 34􀆰 2±4􀆰 18b 32􀆰 69 56􀆰 25 64􀆰 42
C􀆰 himalaica 18􀆰 7±0􀆰 32b 27􀆰 5±0􀆰 73a 27􀆰 5±3􀆰 46a∗ 26􀆰 00±0􀆰 51a∗ 47􀆰 06 47􀆰 06 39􀆰 04
PE A􀆰 thaliana 8􀆰 00±1􀆰 04a 4􀆰 27±0􀆰 75b 4􀆰 31±0􀆰 26b 4􀆰 00±1􀆰 4b -46􀆰 63 -46􀆰 13 -50􀆰 00
C􀆰 himalaica 4􀆰 38±0􀆰 66ab∗ 3􀆰 72±0􀆰 05b 5􀆰 04±0􀆰 81a 4􀆰 77±0􀆰 70a -15􀆰 07 15􀆰 07 8􀆰 90
PI A􀆰 thaliana 5􀆰 43±0􀆰 38b 6􀆰 03±1􀆰 05b 7􀆰 07±0􀆰 62a 7􀆰 28±0􀆰 75a 11􀆰 05 30􀆰 20 34􀆰 07
C􀆰 himalaica 4􀆰 23±0􀆰 05b∗ 4􀆰 87±0􀆰 45ab∗ 5􀆰 56±0􀆰 32a∗ 5􀆰 57±0􀆰 11a∗ 15􀆰 13 31􀆰 44 31􀆰 67
PS A􀆰 thaliana 0􀆰 78±0􀆰 17b 0􀆰 84±0􀆰 25b 1􀆰 07±0􀆰 03a 0􀆰 80±0􀆰 06b 7􀆰 69 31􀆰 18 2􀆰 56
C􀆰 himalaica 0􀆰 79±0􀆰 03a 0􀆰 98±0􀆰 18a 0􀆰 79±0􀆰 2a∗ 0􀆰 86±0􀆰 04a 24􀆰 05 0􀆰 00 8􀆰 86
PA A􀆰 thaliana 0􀆰 38±0􀆰 03b 0􀆰 38±0􀆰 14b 0􀆰 44±0􀆰 03b 0􀆰 86±0􀆰 37a 0􀆰 00 15􀆰 79 126􀆰 32
C􀆰 himalaica 0􀆰 22±0􀆰 02a 0􀆰 18±0􀆰 07a∗ 0􀆰 14±0􀆰 02a∗ 0􀆰 28±0􀆰 05a∗ -18􀆰 18 -36􀆰 36 27􀆰 27
Total A􀆰 thaliana 299􀆰 1±21􀆰 01b 303􀆰 4±69􀆰 82b 405􀆰 9±31􀆰 31a 391􀆰 1±32􀆰 35a 1􀆰 44 35􀆰 71 30􀆰 76
C􀆰 himalaica 222􀆰 9±10􀆰 22b∗ 324􀆰 3±17􀆰 85a 336􀆰 9±32􀆰 28a∗ 312􀆰 9±23􀆰 36a∗ 45􀆰 49 51􀆰 14 40􀆰 38
Table 2  Total amount of lipid in each head ̄class group and total lipids in A􀆰 thaliana and C􀆰 himalaica roots under different K+ levels.
Data are means ± SD (n= 4 or 5) . Values in the same row with different letters differ significantly at P < 0􀆰 05 (Fisher’s least significant
difference) . Asterisk indicates that the value is different from that of A􀆰 thaliana under the same conditions (P < 0􀆰 05)
Lipid
class
  Plant
  species
Lipid / nmol􀅰mg-1 DW
Control   
0􀆰 51
mmol􀅰L-1
0􀆰 051
mmol􀅰L-1
  0
  mmol􀅰L-1
Relative change / %
0􀆰 51
mmol􀅰L-1
0􀆰 051
mmol􀅰L-1

mmol􀅰L-1
DGDG A􀆰 thaliana 1􀆰 44±0􀆰 13b 1􀆰 43±0􀆰 23b 1􀆰 86±0􀆰 35a 1􀆰 84±0􀆰 3a -0􀆰 69 29􀆰 17 27􀆰 78
C􀆰 himalaica 1􀆰 34±0􀆰 36b 1􀆰 34±0􀆰 14b 1􀆰 79±0􀆰 36a 1􀆰 67±0􀆰 34ab 0􀆰 00 33􀆰 58 24􀆰 63
MGDG A􀆰 thaliana 4􀆰 67±0􀆰 34a 3􀆰 86±1􀆰 77a 5􀆰 21±0􀆰 55a 5􀆰 38±0􀆰 80a -17􀆰 34 11􀆰 56 15􀆰 20
C􀆰 himalaica 3􀆰 76±1􀆰 00a 2􀆰 67±1􀆰 26a 3􀆰 95±1􀆰 49a 3􀆰 42±1􀆰 60a∗ -28􀆰 99 5􀆰 05 -9􀆰 04
PG A􀆰 thaliana 1􀆰 88±0􀆰 23a 1􀆰 44±0􀆰 32b 1􀆰 74±0􀆰 15a 1􀆰 31±0􀆰 10b -23􀆰 40 -7􀆰 45 -30􀆰 32
C􀆰 himalaica 1􀆰 53±0􀆰 25a∗ 1􀆰 36±0􀆰 11a 1􀆰 54±0􀆰 04a 1􀆰 62±0􀆰 32a∗ -11􀆰 11 0􀆰 65 5􀆰 88
PC A􀆰 thaliana 26􀆰 9±3􀆰 01ab 29􀆰 6±9􀆰 68a 30􀆰 4±2􀆰 37a 24􀆰 1±2􀆰 08b 10􀆰 04 13􀆰 01 -10􀆰 41
C􀆰 himalaica 22􀆰 4±2􀆰 48a 23􀆰 8±2􀆰 12a∗ 22􀆰 5±1􀆰 88a∗ 24􀆰 8±3􀆰 23a 6􀆰 25 0􀆰 45 10􀆰 71
PE A􀆰 thaliana 6􀆰 83±0􀆰 84a 5􀆰 98±1􀆰 89a 5􀆰 81±0􀆰 36ab 4􀆰 71±0􀆰 2b -12􀆰 45 -14􀆰 93 -31􀆰 04
C􀆰 himalaica 5􀆰 35±0􀆰 92a∗ 5􀆰 06±0􀆰 11a 4􀆰 35±0􀆰 17a∗ 4􀆰 58±0􀆰 54a -5􀆰 42 -18􀆰 69 -14􀆰 39
PI A􀆰 thaliana 5􀆰 83±0􀆰 47a 6􀆰 51±2􀆰 16a 6􀆰 48±0􀆰 55a 5􀆰 22±0􀆰 17a 11􀆰 66 11􀆰 15 -10􀆰 46
C􀆰 himalaica 5􀆰 11±0􀆰 98a 5􀆰 28±0􀆰 37a 5􀆰 00±0􀆰 71a∗ 5􀆰 40±0􀆰 15a 3􀆰 33 -2􀆰 15 5􀆰 68
PS A􀆰 thaliana 0􀆰 96±0􀆰 18a 1􀆰 04±0􀆰 44a 0􀆰 99±0􀆰 16a 0􀆰 59±0􀆰 06b 8􀆰 33 3􀆰 13 -38􀆰 54
C􀆰 himalaica 0􀆰 92±0􀆰 22ab 1􀆰 11±0􀆰 1a 0􀆰 61±0􀆰 05c∗ 0􀆰 68±0􀆰 04bc 20􀆰 65 -33􀆰 70 -26􀆰 09
PA A􀆰 thaliana 2􀆰 23±0􀆰 45b 3􀆰 67±1􀆰 13a 3􀆰 57±0􀆰 35a 3􀆰 54±0􀆰 70a 64􀆰 57 60􀆰 09 58􀆰 74
C􀆰 himalaica 3􀆰 13±0􀆰 59b 4􀆰 24±0􀆰 36a 2􀆰 79±0􀆰 73b 3􀆰 41±0􀆰 88ab 35􀆰 46 -10􀆰 86 8􀆰 95
Total A􀆰 thaliana 51􀆰 2±4􀆰 63ab 53􀆰 9±15􀆰 83ab 56􀆰 6±4􀆰 07a 47􀆰 5±2􀆰 93b 5􀆰 27 10􀆰 55 -7􀆰 23
C􀆰 himalaica 43􀆰 8±4􀆰 82a 44􀆰 9±2􀆰 63a∗ 42􀆰 6±3􀆰 99a∗ 46􀆰 2±4􀆰 52a 2􀆰 51 -2􀆰 74 5􀆰 48
071                                  植 物 分 类 与 资 源 学 报                            第 36卷
increased (Table 3). Howeverꎬ no significant change
was observed in the relative levels of GL and PL in
K+ ̄deficient A􀆰 thaliana leaves (Table 4). The pat ̄
terns of change in molecular species of DGDGꎬ MG ̄
DG and PG were consistent with the overall changes
in DGDGꎬ MGDG and PG classesꎻ almost all molec ̄
ular species of these three lipid classes showed in ̄
creases after K+  ̄deficient treatment (Fig􀆰 4A).
Extraplastidic lipidsꎬ such as PCꎬ PI and PSꎬ
showed increases under K+ ̄deficient conditions (Table
1). This was most pronounced for PCꎬ which increased
by 32􀆰 69%ꎬ 56􀆰 25% and 54􀆰 42% under 0􀆰 51ꎬ 0􀆰 051
and 0mmol􀅰L-1 K+ conditionsꎬ respectivelyꎬ compared
with the control. Almost all molecular species in the
categories of PCꎬ PI and PS showed increasesꎬ except
for 34 ∶ 2 and 42 ∶ 2 PSꎬ which decreased (Fig􀆰 4A). As
a structural component of membranesꎬ it is also well
known that PA acts as a cellular messenger and its lev ̄
el can increase rapidly in plants under various stress
conditions. Under 0􀆰 051 and 0 mmol􀅰L-1 K+ condi ̄
tionsꎬ PA increased by 15􀆰 79% and 126􀆰 32%ꎬ re ̄
spectivelyꎬ compared with the control ( Table 1).
Indeedꎬ the levels of all molecular species of PA in ̄
creased under K+  ̄limited conditions (Fig􀆰 4A).
The molecular shape of PA is a coneꎬ and it
has a strong propensity to form a hexagonal II phase
(Verkleij et al.ꎬ 1982)ꎬ a non ̄lamellar phase that
has a strong tendency to cause membrane leakage. Pre ̄
vious studies support a negative relationship between
the level of PA and the tolerance of plants to freezing
(Li et al.ꎬ 2008ꎻ Welti et al.ꎬ 2002ꎻ Zhang et al.ꎬ
2012). The increase in PA in A􀆰 thaliana leaves may
be responsible for impairing membrane stabilityꎬ resul ̄
ting in severe necrosis in aged leaves. Different from
PCꎬ PIꎬ PS and PAꎬ the amount of PE dramatically
decreased under K+ ̄deficient conditionsꎬ by 46􀆰 63%ꎬ
Table 3  Amount of PL and GL in A􀆰 thaliana and C􀆰 himalaica under different K+ levels. Data are means ± SD (n= 4 or 5) .
Values in the same row with different letters differ significantly at P < 0􀆰 05 (Fisher’s least significant difference) .
Asterisk indicates that the value is different from that of A􀆰 thaliana under the same conditions (P < 0􀆰 05)
Lipid
class
  Plant
  species
Lipid / nmol􀅰mg-1 DW
Control   
0􀆰 51 
mmol􀅰L-1  
0􀆰 051
mmol􀅰L-1
  0
  mmol􀅰L-1
Relative change / %
0􀆰 51
mmol􀅰L-1
0􀆰 051
mmol􀅰L-1

mmol􀅰L-1
PL of L A􀆰 thaliana 54􀆰 4±3􀆰 45b 61􀆰 6±2􀆰 69ab 68􀆰 8±3􀆰 43a 62􀆰 7±17􀆰 6ab 13􀆰 24 26􀆰 47 15􀆰 26
C􀆰 himalaica 37􀆰 4±0􀆰 79b∗ 53􀆰 8±1􀆰 96a 54􀆰 2±5􀆰 96a∗ 51􀆰 1±1􀆰 06a∗ 43􀆰 85 44􀆰 92 36􀆰 63
PL of R A􀆰 thaliana 45􀆰 1±4􀆰 3ab 48􀆰 6±15􀆰 5a 49􀆰 5±3􀆰 46a 39􀆰 1±0􀆰 64b 7􀆰 76 9􀆰 76 -13􀆰 30
C􀆰 himalaica 38􀆰 7±4􀆰 21a 42􀆰 0±0􀆰 9a 36􀆰 8±3􀆰 37a∗ 39􀆰 2±1􀆰 52a 8􀆰 53 -4􀆰 91 1􀆰 29
GL of L A􀆰 thaliana 245􀆰 6±20􀆰 77b 246􀆰 5±64􀆰 24b 337􀆰 2±28􀆰 3a 329􀆰 0±24􀆰 57a 0􀆰 37 37􀆰 30 33􀆰 96
C􀆰 himalaica 185􀆰 5±10􀆰 08b∗ 274􀆰 1±10􀆰 55a 282􀆰 7±32􀆰 23a∗ 260􀆰 1±21􀆰 67a∗ 47􀆰 76 52􀆰 40 40􀆰 22
GL of R A􀆰 thaliana 6􀆰 11±0􀆰 43ab 5􀆰 29±1􀆰 86b 7􀆰 08±0􀆰 77a 7􀆰 21±1􀆰 08a -13􀆰 42 15􀆰 88 18􀆰 00
C􀆰 himalaica 5􀆰 10±1􀆰 35a 4􀆰 00±1􀆰 29a 5􀆰 74±1􀆰 8a 5􀆰 09±1􀆰 67∗ -21􀆰 57 12􀆰 55 -0􀆰 20
Table 4  Relative levels (mol%) of PL and GL in A􀆰 thaliana and C􀆰 himalaica under different K+ levels. Data are means ± SD
(n= 4 or 5) . Values in the same row with different letters differ significantly at P < 0􀆰 05 (Fisher’s least significant difference) .
Asterisk indicates that the value is different from that of A􀆰 thaliana under the same conditions (P < 0􀆰 05)
Lipid
class
  Plant
  species
Lipid / nmol􀅰mg-1 DW
Control     
0􀆰 51 
mmol􀅰L-1  
0􀆰 051
mmol􀅰L-1
  0
  mmol􀅰L-1
Relative change / %
0􀆰 51
mmol􀅰L-1
0􀆰 051
mmol􀅰L-1

mmol􀅰L-1
PL of L A􀆰 thaliana 17􀆰 9±1􀆰 48a 17􀆰 2±0􀆰 34a 16􀆰 7±0􀆰 2a 19􀆰 1±8􀆰 36a   -3􀆰 91   -6􀆰 70 6􀆰 70
C􀆰 himalaica 17􀆰 3±0􀆰 34a 16􀆰 2±0􀆰 54a 16􀆰 2±2􀆰 14a 16􀆰 9±1􀆰 19a   -6􀆰 36   -6􀆰 36 -2􀆰 31
PL of R A􀆰 thaliana 88􀆰 3±0􀆰 18ab 89􀆰 5±4􀆰 86a 87􀆰 5±0􀆰 77ab 84􀆰 8±2􀆰 21b   1􀆰 36   -0􀆰 91 -3􀆰 96
C􀆰 himalaica 88􀆰 4±2􀆰 62ab 91􀆰 1±2􀆰 79a 86􀆰 6±3􀆰 54b 89􀆰 0±3􀆰 39ab∗   3􀆰 05   -2􀆰 04 0􀆰 68
GL of L A􀆰 thaliana 82􀆰 1±1􀆰 48a 82􀆰 8±0􀆰 34a 83􀆰 3±0􀆰 20a 80􀆰 9±8􀆰 36a   0􀆰 85   1􀆰 46 -1􀆰 46
C􀆰 himalaica 82􀆰 7±0􀆰 34a 83􀆰 8±0􀆰 54a 83􀆰 8±2􀆰 14a 83􀆰 1±1􀆰 19a   1􀆰 33   1􀆰 33 0􀆰 48
GL of R A􀆰 thaliana 11􀆰 7±0􀆰 18ab 10􀆰 5±4􀆰 86b 12􀆰 5±0􀆰 77ab 15􀆰 2±2􀆰 21a -10􀆰 3   6􀆰 84 29􀆰 9 
C􀆰 himalaica 11􀆰 6±2􀆰 62ab 8􀆰 93±2􀆰 79b 13􀆰 4±3􀆰 54a 11􀆰 0±3􀆰 39ab∗ -23􀆰 0 15􀆰 5 -5􀆰 17
1712期    WANG Dan ̄Dan et al.: Plants Adapt to Long ̄Term Potassium Deficiency by Accumulation of Membrane 􀆺    
46􀆰 13% and 50% under 0􀆰 51ꎬ 0􀆰 051 and 0 mmol􀅰L-1
K+ conditionsꎬ respectivelyꎬ compared with the control
(Table 1). All molecular species of PE also showed
decreases under K+  ̄limited conditions (Fig􀆰 4A).
Fig􀆰 4  Changes in lipid molecular species in A􀆰 thaliana and C􀆰 himalaica under different K+ conditions
A. Changes in lipid molecular species in A􀆰 thaliana leaves under different K+ conditionsꎻ B. Changes in lipid molecular species in C􀆰 himalaica
leaves under different K+ conditionsꎻ C. Changes in lipid molecular species in A􀆰 thaliana roots under different K+ conditionsꎻ D. Changes in
lipid molecular species in C􀆰 himalaica roots under different K+ conditionsꎻ E. Changes in molecular species of PA and PE in A􀆰 thaliana
leaves under different K+ conditionsꎻ F. Changes in molecular species of PA and PE in A􀆰 thaliana roots under different K+ conditions
271                                  植 物 分 类 与 资 源 学 报                            第 36卷
  Long ̄term nutrient ̄deficiency ̄induced leaf chlo ̄
rosis and necrosis are the consequences of lipid deg ̄
radation in plastidic membranes. Thereforeꎬ it is ver ̄
y interesting that both leaf senescence and increases
in plastidic membrane lipids ( DGDGꎬ MGDG and
PG) were found in long ̄term K+  ̄deficient leaves.
Howeverꎬ further study is needed to clarify the spe ̄
cific mechanisms involved.
2􀆰 4  Lipid changes in A􀆰 thaliana roots differed
from those in leaves
Lipid changes in A􀆰 thaliana roots were small.
The patterns of lipid changes in A􀆰 thaliana roots dif ̄
fered from those of leaves. The changes in the levels
of lipids in roots cannot be simply described as an
increase or a decrease. Insteadꎬ the levels fluctuated
and varied slightly under different K+ conditions. For
exampleꎬ under conditions of 0􀆰 51 mmol􀅰L-1 K+ꎬ
MGDG decreased by 17􀆰 34%ꎬ but it increased by
11􀆰 56% and 15􀆰 2% under 0􀆰 051 and 0 mmol􀅰L-1
K+ conditionsꎬ respectivelyꎬ compared with the con ̄
trol (Table 2). Under K+  ̄limited conditionsꎬ the a ̄
mounts of most lipid classes and total lipids in K+  ̄
deficient roots showed no significant difference com ̄
pared with the control.
There were no clear patterns of change in the
molecular species of lipids of MGDGꎬ PCꎬ PEꎬ PG
and PI in A􀆰 thaliana roots under K+  ̄deficient condi ̄
tions. Most molecular species in the category of PS
showed a decrease under K+  ̄deficient conditionsꎬ
whereas most of DGDG showed an increase (Fig􀆰 4C).
The changes of PA in roots were similar to those in
leaves: it significantly increased in K+  ̄limited
A􀆰 thaliana roots under 0􀆰 51ꎬ 0􀆰 051 and 0 mmol􀅰L-1
K+ conditionsꎬ by 64􀆰 57%ꎬ 60􀆰 09% and 58􀆰 74%ꎬ
respectively (Table 2). The levels of all molecular
species in the category of PA increased significantly
under K+  ̄limited conditions (Fig􀆰 4C).
Under K+  ̄deficient conditionsꎬ changes in root
GL and PL were minor and no significant differences
were observed when compared with the control (Ta ̄
ble 3). The relative levels of GL and PL in K+  ̄
stressed roots showed no significant differences when
compared with the control ( Table 4). Considering
that a previous study reported that the relative levels
of PL did not change with 0ꎬ 24ꎬ 30 and 36 h of K+
deprivation in barley roots (Hafsi et al.ꎬ 2009)ꎬ it
is suggested that plants might adapt to potassium de ̄
ficiency by maintaining the PL composition in roots.
2􀆰 5  Lipid changes in C􀆰 himalaica differed from
those of A􀆰 thaliana
The evidence outlined above for A􀆰 thaliana in ̄
dicates that membrane lipids accumulate in leaves
and are maintained in roots after long ̄term K+ defi ̄
ciency. Given that the level of tolerance to long ̄term
potassium deficiency in C􀆰 himalaica was higher than
that in A􀆰 thalianaꎬ if the pattern of lipid changes in
A􀆰 thaliana is a consequence of plants adapting to
long ̄term potassium deficiencyꎬ the extent of the
increase in lipids will decrease over time in
C􀆰 himalaica. If the pattern of lipid changes is an ad ̄
aptation of plants to long ̄term potassium deficiencyꎬ
the extent of increase in lipids will be maintained or
even increase over time in C􀆰 himalaica.
Under K+  ̄sufficient conditionsꎬ the amounts of
lipids in each head ̄group class and total lipids in
C􀆰 himalaica leaves and roots were less than those in
A􀆰 thaliana (Tables 1 and 2)ꎬ with the exception of
PA in C􀆰 himalaica roots. Under K+  ̄limited condi ̄
tionsꎬ almost all lipid classes in C. himalaica leaves
showed increases after K+  ̄deficient treatment. The
pattern of response of the lipid composition in C. hi ̄
malaica leaves was similar to that of A􀆰 thaliana.
Howeverꎬ the extent of change in C􀆰 himalaica dif ̄
fered from that of A􀆰 thaliana. The extents of the in ̄
creases of DGDGꎬ MGDGꎬ PG and total lipids in
C􀆰 himalaica were larger than those in A􀆰 thaliana.
For exampleꎬ under 0􀆰 51ꎬ 0􀆰 051 and 0 mmol􀅰L-1
K+ conditionsꎬ PG in C􀆰 himalaica increased by
85􀆰 55%ꎬ 69􀆰 27% and 60􀆰 78% respectivelyꎬ com ̄
pared with the control. Meanwhileꎬ A􀆰 thaliana exhibi ̄
ted 15􀆰 46%ꎬ 24􀆰 64% and 6􀆰 66% increases under
the same K+ conditions (Table 1). Thusꎬ the tend ̄
ency for an increase in plastidic lipids in K+  ̄defi ̄
cient C. himalica leaves was relatively enhanced.
3712期    WANG Dan ̄Dan et al.: Plants Adapt to Long ̄Term Potassium Deficiency by Accumulation of Membrane 􀆺    
K+ limitation led to a significant decrease in
A􀆰 thaliana PEꎬ but in K+  ̄deficient C􀆰 himalaica
leavesꎬ the levels of PE showed no difference com ̄
pared with the control. PA levels did tend to be low ̄
er in C􀆰 himalaica leaves than in A􀆰 thaliana under
both control and K+  ̄limited conditions. PA in
A􀆰 thaliana increased by 15􀆰 79% and 126􀆰 32% un ̄
der 0􀆰 051 and 0 mmol􀅰L-1 K+ conditionsꎬ respec ̄
tivelyꎬ whereas in C􀆰 himalaicaꎬ it decreased by
18􀆰 18% and 36􀆰 36% under 0􀆰 51 and 0􀆰 051 mmol
􀅰L-1 K+ conditionsꎬ and under K+  ̄deprivation condi ̄
tionsꎬ it showed a 27􀆰 27% increase (Table 1). How ̄
everꎬ these changes did not constitute significant
differences compared with the control. Almost all
species of DGDGꎬ MGDGꎬ PCꎬ PG and PI in
C􀆰 himalaica leaves showed increases under K+  ̄lim ̄
ited conditions. The levels of most molecular species
in the category of PS increased after K+ ̄deficient
treatmentꎬ but 42 ∶ 2ꎬ 42 ∶ 3ꎬ 44 ∶ 2 and 44 ∶ 3 PS
showed decreases. In additionꎬ the levels of molecu ̄
lar species in the categories of PA and PE changed
slightly under K+  ̄limited conditions (Fig􀆰 4B).
At the root levelꎬ the extent of lipid changes in
C􀆰 himalaica was less than that in A􀆰 thaliana. Under
0􀆰 051 and 0 mmol􀅰L-1 K+ conditionsꎬ the most dra ̄
matic changes in lipid levels occurred in DGDG and
PS. DGDG showed 33􀆰 58% and 24􀆰 63% increases
under 0􀆰 051 and 0 mmol􀅰L-1 K+ conditionsꎬ respec ̄
tivelyꎬ whereas the amount of PS decreased by
33􀆰 7% and 26􀆰 09% under the same K+ conditions
(Table 2). PA content in A􀆰 thaliana leaves was sig ̄
nificantly higher than that in C􀆰 himalaica under both
K+ ̄sufficient and K+ ̄deficient conditions. At the root
levelꎬ C􀆰 himalaica produced more PA than A􀆰 thali ̄
ana under 5􀆰 1 and 0􀆰 51 mmol􀅰L-1 K+ conditionsꎬ
whereas when grown under 0􀆰 051 and 0 mmol􀅰L-1
K+ conditionsꎬ PA levels were higher in A􀆰 thalianaꎬ
but no significant differences were found between
these two species under the four levels of potassium
exposure (Table 2). We speculate that the increased
levels of PA in A􀆰 thaliana membranes under K+ ̄defi ̄
cient conditions promote non ̄bilayer phase formation
that leads to cell death. The lower PA in C􀆰 himalaica
under K+ ̄deficient conditions would thus be the basis
for the better growth of C􀆰 himalaica.
C􀆰 himalaica exhibited greater increases in GL
and PL than A􀆰 thaliana under the same K+  ̄deficient
conditions ( Table 3). The relative contents of GL
and PL in K+  ̄stressed C􀆰 himalaica leaves and roots
showed no significant changes when compared with
the control (Table 4). The levels of most molecular
species in C􀆰 himalaica roots were maintained or
changed only slightly under K+  ̄limited conditionsꎬ
with the exception of PS. Most molecular species in
the category of PS exhibited a decrease after K+  ̄de ̄
ficient treatmentꎬ except for 36 ∶ 4ꎬ 38 ∶ 2ꎬ 38 ∶ 3 and
42 ∶ 4ꎬ the levels of the first three slightly increasedꎬ
while the fourth was maintained under K+  ̄limited
conditions (Fig􀆰 4D).
The foregoing analysis indicates thatꎬ under K+ ̄
deficient conditionsꎬ the extent of increase in plastid ̄
ic lipids in C􀆰 himalaica leaves was larger than that
in A􀆰 thalianaꎬ whereas lipid changes in roots were
less than those in A􀆰 thaliana. These results suggest
that the tendency for a change in lipids in C􀆰 himalaica
was enhanced. This result is consistent with our pre ̄
vious hypothesisꎬ namelyꎬ the pattern of lipid change
in A􀆰 thaliana and C􀆰 himalaica is an active adapta ̄
tion of plants to long ̄term potassium deficiencyꎬ
rather than simply an indirect consequence of plants
adapting to long ̄term potassium deficiency.
2􀆰 6  The increases of PA species were associated
with the decreases of PE species in K+ ̄stressed
A􀆰 thaliana
We found thatꎬ in A􀆰 thaliana leavesꎬ 34 ∶ 2ꎬ
34 ∶ 3ꎬ 36 ∶ 4ꎬ 36 ∶ 5 and 36 ∶ 6 PA significantly in ̄
creased under K+  ̄deficient conditionsꎬ while 34 ∶ 2ꎬ
34 ∶ 3ꎬ 36 ∶ 4ꎬ 36 ∶ 5 and 36 ∶ 6 PE decreased corre ̄
spondingly (Fig􀆰 4A and 4E). This indicates that K+ ̄
deficiency ̄induced PA in leaves of A􀆰 thaliana was de ̄
rived mainly from PE. Howeverꎬ in C􀆰 himalaicaꎬ
this relationship between PA and PE was not ob ̄
served. The levels of almost all molecular species of
PA showed decreases under K+  ̄deficient conditionsꎬ
471                                  植 物 分 类 与 资 源 学 报                            第 36卷
except for 34 ∶ 3ꎬ 36 ∶ 5 and 36 ∶ 6 PAꎬ which exhibi ̄
ted increases at 0 mmol􀅰L-1 K+ . Meanwhileꎬ the
levels of PE species were maintained or increased
under K+  ̄deficient conditionsꎬ with the exception of
34 ∶ 2 PEꎬ which showed a decrease (Fig􀆰 4B).
The changes in lipids in K+  ̄stressed roots were
more complex than those in leaves. Under K+  ̄defi ̄
cient conditionsꎬ the levels of DGDGꎬ MGDG and
PA species increased in A􀆰 thaliana rootsꎬ whereas
those of most PEꎬ PG and PS species decreased.
The contents of most PC and PI species remained at
the control level (Fig􀆰 4C). Under K+  ̄deficient con ̄
ditionsꎬ most lipid species in C􀆰 himalaica roots were
maintained or increased slightly when compared with
the controlꎬ with the exception of 34 ∶ 2 and 34 ∶ 3 of
PE and 34 ∶ 2ꎬ 34 ∶ 3ꎬ 40 ∶ 2ꎬ 40 ∶ 3ꎬ 42 ∶ 2 and 42 ∶ 3
of PSꎬ which showed decreases under K+  ̄limited
conditions ( Fig􀆰 4D ). PA levels in the roots of
A􀆰 thaliana and C􀆰 himalaica were significantly high ̄
er than those in leaves under both control and K+  ̄
limited conditions (Table 2). K+ deficiency led to a
large increase in PA species in roots of A􀆰 thaliana
(Fig􀆰 4C). In these rootsꎬ there was also a correla ̄
tion between PA and PE (Fig􀆰 4C and 4F). Howev ̄
erꎬ the PA levels in the roots of K+  ̄stressed
C􀆰 himalaica showed no significant change when
compared with the controlꎬ and the levels of 34 ∶ 3
and 36 ∶ 3 of PA actually decreased (Fig􀆰 4D).
3  Conclusion
Our results confirm that a relative of A􀆰 thalianaꎬ
C􀆰 himalaicaꎬ is tolerant to low ̄K+ stress. Less chlo ̄
rosis and a smaller reduction in total potassium con ̄
tent in C􀆰 himalaica under K+  ̄limited conditions
suggested that A􀆰 thaliana is far more sensitive to low
K+ than C􀆰 himalaica. Under K+ ̄deficient conditionsꎬ
the R ∶ S dry weight ratio decreased in A􀆰 thalianaꎬ
whereas it significantly increased in C􀆰 himalaica.
We suggest that this is the one of the factors that ac ̄
count for the higher level of potassium and better
growth of C􀆰 himalaica under K+  ̄deficient condi ̄
tions. In additionꎬ in K+  ̄stressed C􀆰 himalaicaꎬ a
considerable proportion of potassium was allocated to
the leaves. This may be responsible for the lower lev ̄
el of chlorosis of C􀆰 himalaica under K+  ̄deficient
conditions.
The results of lipid composition analysis indi ̄
cate that plant leaves and roots respond differently to
K+ deficiency. The levels of most lipid classes in ̄
creased in shoots of A􀆰 thaliana and C􀆰 himalaicaꎬ
whereas fewer changes in lipids occurred in response
to K+ deficiency in the roots of these species. This
indicated that plants adapt to long ̄term K+ by the
accumulation of lipids in their leaves and by the ma ̄
intenance of lipid composition in their roots. The
structure of PA promotes the tendency to form a hex ̄
agonal II phaseꎬ which has a negative effect on
membrane integrity. We propose that less PA accu ̄
mulation in C􀆰 himalaica is another important factor
that explains why it can maintain better growth under
K+  ̄deficient conditions. Detailed analysis of the pro ̄
files of lipid molecular species also indicates that
K+  ̄deficiency ̄induced PA in A􀆰 thaliana is derived
primarily from PE.
Acknowledgments: The authors thank Mary Roth (Kansas
Lipidomics Research Center) for lipid analysis.
References:
Al ̄Shehbaz IAꎬ Beilstein MAꎬ Kellogg EAꎬ 2006. Systematics and
phylogeny of the Brassicaceae ( Cruciferae): an overview [ J] .
Plant Systematics and Evolutionꎬ 259 (2): 89—120
Armengaud Pꎬ Breitling Rꎬ Amtmann Aꎬ 2004. The potassium ̄de ̄
pendent transcriptome of Arabidopsis reveals a prominent role of
jasmonic acid in nutrient signaling [ J] . Plant Physiologyꎬ 136
(1): 2556—2576
Armengaud Pꎬ Sulpice Rꎬ Miller AJ et al.ꎬ 2009. Multilevel analysis
of primary metabolism provides new insights into the role of po ̄
tassium nutrition for glycolysis and nitrogen assimilation in Arabi ̄
dopsis roots [J] . Plant Physiologyꎬ 150 (2): 772—785
Ashley MKꎬ Grant Mꎬ Grabov Aꎬ 2006. Plant responses to potassium
deficiencies: a role for potassium transport proteins [J] . Journal
of Experimental Botanyꎬ 57 (2): 425—436
Cakmak Iꎬ Hengeler Cꎬ Marschner Hꎬ 1994. Partitioning of shoot and
root dry matter and carbohydrates in bean plants suffering from
phosphorusꎬ potassium and magnesium deficiency [J] . Journal
of Experimental Botanyꎬ 45 (9): 1245—1250
5712期    WANG Dan ̄Dan et al.: Plants Adapt to Long ̄Term Potassium Deficiency by Accumulation of Membrane 􀆺    
Chen JPꎬ Burke JJꎬ Xin ZG et al.ꎬ 2006. Characterization of the Ara ̄
bidopsis thermosensitive mutant atts02 reveals an important role
for galactolipids in thermotolerance [J] . Plantꎬ Cell and Envi ̄
ronmentꎬ 29 (7): 1437—1448
Chen YFꎬ Wang Yꎬ Wu WHꎬ 2008. Membrane transporters for nitro ̄
genꎬ phosphate and potassium uptake in plants [J] . Journal of
Integrative Plant Biologyꎬ 50 (7): 835—848
Gierth Mꎬ Mäser Pꎬ Schroeder JIꎬ 2005. The potassium transporter
AtHAK5 functions in K+ deprivation ̄induced high ̄affinity K+ up ̄
take and AKT1 K+ channel contribution to K+ uptake kinetics in
Arabidopsis roots [J] . Plant Physiologyꎬ 137 (3): 1105—1114
Guan LZ (关连珠)ꎬ 2001. Soil fertilizer science (土壤肥料学)
[M]. Beijing: Chinese Agricultural Pressꎬ 138
Hafsi Cꎬ Russo MAꎬ Sgherri C et al.ꎬ 2009. Implication of phospho ̄
lipase D in response of Hordeum vulgare root to short ̄term potassi ̄
um deprivation [ J] . Journal of Plant Physiologyꎬ 166 ( 5):
499—506
Hermans Cꎬ Hammond JPꎬ White PJ et al.ꎬ 2006. How do plants re ̄
spond to nutrient shortage by biomass allocation? [ J] . Trends in
Plant Scienceꎬ 11 (12): 610—617
Hong YYꎬ Devaiah SPꎬ Bahn SC et al.ꎬ 2009. Phospholipase Dε and
phosphatidic acid enhance Arabidopsis nitrogen signaling and
growth [J] . The Plant Journalꎬ 58 (3): 376—387
Hong YYꎬ Pan XQꎬ Welti R et al.ꎬ 2008. Phospholipase Dα3 is in ̄
volved in the hyperosmotic response in Arabidopsis [ J] . Plant
Cellꎬ 20 (3): 803—816
Hong YYꎬ Zhang WHꎬ Wang XMꎬ 2010. Phospholipase D and phos ̄
phatidic acid signalling in plant response to drought and salinity
[J] . Plantꎬ Cell and Environmentꎬ 33 (4): 627—635
Leigh RAꎬ Wyn Jones RGꎬ 1984. A hypothesis relating critical potas ̄
siun concentrations for growth to the distribution and functions of
this ion in the plant cell [J] . New Phytologistꎬ 97 (1): 1—13
Li HW (李宏伟)ꎬ Jiang CXL (江村西罗)ꎬ Li H (李琥) et al.ꎬ
eds. 2003. Baima Xueshan National Nature Reserve (白马雪山
国家级自然保护区) [M]. Kunming: The Nationalities Pub ̄
lishing House of Yunnanꎬ 58
Li MYꎬ Welti Rꎬ Wang XMꎬ 2006a. Quantitative profiling of Arabi ̄
dopsis polar glycerolipids in response to phosphorus starvation.
Roles of Phospholipases Dζ1 and Dζ2 in phosphatidylcholine hy ̄
drolysis and digalactosyldiacylglycerol accumulation in phosphor ̄
us ̄starved plants [J] . Plant Physiologyꎬ 142 (2): 750—761
Li MYꎬ Qin CBꎬ Welti R et al.ꎬ 2006b. Double knockouts of phos ̄
pholipases Dζ1 and Dζ2 in Arabidopsis affect root elongation dur ̄
ing phosphate ̄limited growth but do not affect root hair patterning
[J] . Plant Physiologyꎬ 140 (2): 761—770
Li WQꎬ Li MYꎬ Zhang WH et al.ꎬ 2004. The plasma membrane ̄
bound phospholipase Dδ enhances freezing tolerance in Arabidop ̄
sis thaliana [J] . Nature Biotechnologyꎬ 22 (4): 427—433
Li WQꎬ Wang RPꎬ Li MY et al.ꎬ 2008. Differential degradation of ex ̄
traplastidic and plastidic lipids during freezing and post ̄freezing
recovery in Arabidopsis thaliana [ J ] . Journal of Biological
Chemistryꎬ 283 (1): 461—468
Lu RK (鲁如坤)ꎬ 1989. General status of nutrients (Nꎬ Pꎬ K) in
soils of China [ J] . Acta Pedologica Sinica (土壤学报)ꎬ 26
(3): 280—286
Ma TLꎬ Wu WHꎬ Wang Yꎬ 2012. Transcriptome analysis of rice root
responses to potassium deficiency [J] . BMC Plant Biologyꎬ 12
(1): 161
Marschner Hꎬ 1995. Mineral nutrition of higher plants. London: Aca ̄
demic Pressꎬ 310
Russo MAꎬ Quartacci MFꎬ Izzo R et al.ꎬ 2007. Long ̄ and short ̄term
phosphate deprivation in bean roots: Plasma membrane lipid al ̄
terations and transient stimulation of phospholipases [J] . Phyto ̄
chemistryꎬ 68 (11): 1564—1571
Tocquin Pꎬ Corbesier Lꎬ Havelange A et al.ꎬ 2003. A novel high effi ̄
ciencyꎬ low maintenanceꎬ hydroponic system for synchronous
growth and flowering of Arabidopsis thaliana [J] . BMC Plant Bi ̄
ologyꎬ 3 (2): 1—10
Verkleij AJꎬ de Maagd Rꎬ Leunissen ̄Bijvelt J et al.ꎬ 1982. Divalent
cations and chlorpromazine can induce non ̄bilayer structures in
phosphatidic acid ̄containing model membranes [ J] . Biochimica
et Biophysica Acta (BBA)-Biomembranesꎬ 684 (2): 255—262
Wang Cꎬ Chen HFꎬ Hao QN et al.ꎬ 2012. Transcript profile of the re ̄
sponse of two soybean genotypes to potassium deficiency [ J] .
PLoS ONEꎬ 7 (7): e39856
Welti Rꎬ Li WQꎬ Li MY et al.ꎬ 2002. Profiling membrane lipids in
plant stress responses [J] . Journal of Biological Chemistryꎬ 277
(35): 31994—32002
Yang XEꎬ Liu JXꎬ Wang WM et al.ꎬ 2004. Potassium internal use ef ̄
ficiency relative to growth vigorꎬ potassium distributionꎬ and
carbohydrate allocation in rice genotypes [ J] . Journal of Plant
Nutritionꎬ 27 (5): 837—852
Zhang XDꎬ Wang RPꎬ Zhang FJ et al.ꎬ 2012. Lipid profiling and toler ̄
ance to low ̄temperature stress in Thellungiella salsuginea in com ̄
parison with Arabidopsis thaliana [J]. Biologia Plantarumꎬ 1—5
Zhao Jꎬ Wang CXꎬ Bedair M et al.ꎬ 2011. Suppression of phospho ̄
lipase Dγs confers increased aluminum resistance in Arabidopsis
thaliana [J] . PLoS ONEꎬ 6 (12): e28086
Zheng GWꎬ Tian BOꎬ Zhang FJ et al.ꎬ 2011. Plant adaptation to fre ̄
quent alterations between high and low temperatures: remodelling
of membrane lipids and maintenance of unsaturation levels [ J] .
Plantꎬ Cell and Environmentꎬ 34 (9): 1431—1442
671                                  植 物 分 类 与 资 源 学 报                            第 36卷