全 文 :拟南芥中磷脂酶 Dδ参与机械伤害诱导的磷脂酸生成∗
李爱花1ꎬ2ꎬ 凌立贞1ꎬ 李唯奇1∗∗
(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 昆明 650201ꎻ 2 中国科学院大学ꎬ 北京 100049)
摘要: 磷脂酶水解磷脂产生磷脂酸 (phosphatidic acidꎬ PA)ꎬ Dα1和 δ是磷脂酶 D家族中表达丰度最高的
两个成员ꎬ 已知磷脂酶 Dα1参与了机械伤害诱导的磷脂酸信号ꎬ 但是磷脂酶 Dδ 是否以及如何参与 PA 信
号尚且未知ꎮ 本研究利用脂类组学分析方法ꎬ 比较了拟南芥野生型 (WS) 和磷脂酶 Dδ基因 T ̄DNA插入突
变体 (PLDδ ̄KO)ꎬ 在机械伤害后的较长时间段 (6 h) 的膜脂分子变化ꎮ 结果发现ꎬ 机械伤害后ꎬ 拟南芥
两种基因型的大部分膜脂均发生下降ꎬ 且机械伤害后 30 minꎬ PA 含量即快速并急剧升高ꎻ 随着时间的延
长ꎬ 其水平持续升高ꎬ 直至达到峰值后下降至 6 h达到最低值ꎮ WS 和 PLDδ ̄KO 达到 PA 最高值的时间不
同ꎬ 分别为 1 h和 3 hꎻ 在伤害处理后 30 min至 3 h期间ꎬ PLDδ ̄KO中的 PA水平低于 WSꎬ 两个基因型中
的 PA含量最大差值为 20%ꎬ 发生在伤害后 1 hꎮ 这证明缺失 PLDδ基因在一定程度抑制了机械伤害诱导的
PA生产ꎬ 表明 PLDδ参与拟南芥响应机械伤害的 PA生成ꎬ 但是其响应较 PLDα1作用慢且轻ꎮ 这是 PLDδ
响应拟南芥中机械伤害的首次报道ꎮ
关键词: 拟南芥ꎻ 机械伤害ꎻ 膜脂ꎻ 磷脂酶 Dδꎻ 磷脂酸
中图分类号: Q 945 文献标志码: A 文章编号: 2095-0845(2015)04-428-11
Phospholipase Dδ is Involved in Wounding ̄Induced
Phosphatidic Acid Formation in Arabidopsis∗
Li Ai ̄hua1ꎬ2ꎬ Ling Li ̄zhen1ꎬ Li Wei ̄qi1∗∗
( 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 China)
Abstract: Phosphalipase D (PLD) hydrolyzes phospholipids into phosphatidic acid (PA). PLDα1 and δ are the
two most abundant members of the 12 ̄member PLD family in Arabidopsis. PLDα1 has been demonstrated having role
in the wounding ̄induced PA signalling. Howeverꎬ whether and how PLDδ is involved in wounding ̄induced PA for ̄
mation remained unclear. In the present studyꎬ the membrane lipids response to wounding was profiled in Wassil ̄
ewskija (WS) and PLDδ knockout mutant (PLDδ ̄KO) of Arabidopsis. The levels of most lipidsꎬ including mo ̄
nogalactosyldiacylglycerolꎬ digalactosyldiacylglycerolꎬ phosphatidylcholine and phosphatidylglycerol had decreased
rapidly within 30 min after wounding in the two Arabidopsis genotypes. In contrastꎬ the level of PA increased sharply
and significantly 30 min after wounding. It continued to increase until peaking at 1 h post ̄wounding in WS and 3 h
post ̄wounding in PLDδ ̄KOꎬ and then decreased. The PA levels were similar in the two genotypes in untreated leav ̄
es and in leaves 6 h after wounding. Howeverꎬ these levels were lower in PLDδ ̄KO than in WS from 30 min to 3 h
post ̄wounding. The significant difference of PA level between the two genotypes occurred 30 min after woundingꎬ
when it was about 20% lower in PLDδ ̄KO than in WS. These results show that PLDδ is involved in wounding ̄in ̄
duced PA formation in Arabidopsisꎬ but its absence induces PA response later and with less intensity than PLDα1.
植 物 分 类 与 资 源 学 报 2015ꎬ 37 (4): 428~438
Plant Diversity and Resources DOI: 10.7677 / ynzwyj201514058
∗
∗∗
Funding: The National Science Foundation of China (NSFC31300261)
Author for correspondenceꎻ E ̄mail: weiqili@mail kib ac cn
Received date: 2014-04-06ꎬ Accepted date: 2014-06-10
作者简介: 李爱花 (1980-) 女ꎬ 工程师ꎬ 主要从事植物抗逆生理研究ꎮ E ̄mail: liaihua1980@126 com
Key words: Arabidopsisꎻ Woundingꎻ Membrane lipidꎻ PLDδꎻ Phosphatidic acid
Abbreviations: PLDꎬ phospholipase Dꎻ DGDGꎬ digalactosyldiacylglycerolꎻ ESIꎬ electrospray ionisationꎻ MS / MSꎬ
tandem mass spectrometryꎻ MGDGꎬ monogalactosyldiacylglycerolꎻ PAꎬ phosphatidic acidꎻ PCꎬ phosphatidylcho ̄
lineꎻ PEꎬ phosphatidylethanolamineꎻ PGꎬ phosphatidylglycerolꎻ PIꎬ phosphatidylinositolꎻ PSꎬ phosphatidylserineꎻ
ACLꎬ acyl chain lengthꎻ DBIꎬ double ̄bond index
Herbivores constitute a threat to plants by indu ̄
cing wounding and rapidly destroying plant tissues.
Plants can also suffer damage from environmental
stressesꎬ such as windꎬ sandꎬ hail and rainꎬ which
also produce mechanical woundingꎻ this induces
similar responses to those to wounding by herbivores
(Buchanan et al.ꎬ 2004). In response to woundingꎬ
plants adopt both direct and indirect defensive strate ̄
gies (Wasternack et al.ꎬ 2006). Wounding causes
rapid activation of phospholipase D (PLD) ̄mediated
hydrolysis as a result of a rapid accumulation of cho ̄
line and phosphatidic acid (PA) (Ryu and Wangꎬ
1996)ꎻ the latter is used for the synthesis of jasmon ̄
ic acidꎬ a main responder to wounding (Wang et
al.ꎬ 2000). The secondary messenger PA serves a
wide range of signalling roles in plant responses to
environmental stresses (Wangꎬ 2004). PA induc ̄
tion by wounding has been observed in several plant
speciesꎬ including castor bean ( Ryu and Wangꎬ
1996)ꎬ tomatoꎬ soybeanꎬ sunflowerꎬ broad beanꎬ
pepper (Lee et al.ꎬ 1997)ꎬ tobacco (Dhondt et al.ꎬ
2000)ꎬ and Arabidopsis thaliana (Ling et al.ꎬ 2007).
Multiple PLDs have been described in various
plant species (Hanahan and Chaikoffꎬ 1947ꎻ Qin et
al.ꎬ 1997ꎻ Wangꎬ 2000ꎻ Wang et al.ꎬ 1994ꎻ Xu et
al.ꎬ 1997). The 12 PLD genes from Arabidopsis
have been grouped into six classes—PLDα (3)ꎬ β
(2)ꎬ γ (3)ꎬ δꎬ ε and ζ (2)—based on their se ̄
quence similarities and biochemical properties
(Zhang et al.ꎬ 2005). PLDαꎬ PLDβ and PLDγ
have been characterised in response to wounding in
Arabidopsis ( Pappan et al.ꎬ 1998ꎻ Pappan et al.ꎬ
1997aꎬ bꎻ Qin et al.ꎬ 1997ꎻ Wangꎬ 1999). PLDα1
and PLDδ are the two most abundant members of the
Arabidopsis PLD family. Wang et al. (2000) repor ̄
ted that the level of PA increased within 5 min after
woundingꎬ and the maximum difference in the level
of PA between PLDα1 ̄deficient and wild type leaves
occurred 15 min after woundingꎬ when that in the
WT was 2 5 ̄fold higher than that in the PLDα1 ̄de ̄
ficient strain (Wang et al.ꎬ 2000). PLDδ is in ̄
volved in dehydration ̄ and freezing ̄ induced PA for ̄
mation in Arabidopsis (Katagiri et al.ꎬ 2001ꎻ Li et
al.ꎬ 2008). PLDα1 can compensate for the loss of
PLDδ in PLDδ ̄KO mutantꎬ and these two PLD iso ̄
forms together were shown to account for almost all
the discernible activity seen in response to water def ̄
icit ( Bargmann et al.ꎬ 2009). Howeverꎬ whether
and if soꎬ how PLDδ is involved in wounding ̄in ̄
duced PA production remained unknown.
With the aim of determining the role of PLDδ in
lipid metabolism in response to woundingꎬ lipidomic
profiling was carried out for Arabidopsis leaves in the
Wassilewskija (WS) and PLDδ knockout ( PLDδ ̄
KO) genotypesꎬ with and without wounding treat ̄
ment. We used bioinformatics methodsꎬ including
principal component analysis (PCA) and K ̄means /
medians clusteringꎬ to analyse the function of PLDδ
in wounding ̄induced lipid metabolism. Our results
showed that about 20% of PA production upon
wounding was attributable to PLDδ at the time of
statistically significant difference in its level between
WS and PLDδ ̄KO. Moreoverꎬ an absence of PLDδ
slowed the PA response to wounding. The responses
of desaturation and acyl chain length in membrane
lipids to wounding were also discussed.
1 Materials and methods
1 1 Plant materialꎬwound treatment and sampling
Leaves from approximately five ̄week ̄old Arabi ̄
dopsis plants of WS and PLDδ ̄KO in a WS back ̄
ground were used for the identification and kinetic a ̄
9244期 Li Ai ̄hua et al.: Phospholipase Dδ is Involved in Wounding ̄Induced Phosphatidic Acid Formation in
nalysis of lipid species. The loss of PLDδ in the
PLDδ ̄KO mutant of Arabidopsis has been proved by
the absence of its transcriptꎬ protein and activity
(Zhang et al.ꎬ 2003). The plants were cultivated in
a walk ̄in chamber at 20-23 ℃ꎬ with a light intensi ̄
ty of 120 μmol m-2s-1ꎬ a 12 ̄h photoperiod and 60%
relative humidity. In additionꎬ they were fully wa ̄
tered for five weeks until the beginning of the experi ̄
ments. The method of wounding treatment was the
same as that described by Ling et al. (2007).
1 2 Lipid extractionꎬ ESI ̄MS/ MS and data pro ̄
cessing
Lipid extractionꎬ sample analysis and data pro ̄
cessing were performed as described previously with
minor modifications (Li et al.ꎬ 2008ꎻ Welti et al.ꎬ
2002). Brieflyꎬ the leaves were harvested at the
sampling time and transferred immediately into 2 mL
of isopropanol with 0 01% butylated hydroxytoluene
at 75 ℃ . The tissue was extracted with chloroform /
methanol (2 ∶ 1) three additional times with 2 h of
agitation each time. The remaining plant tissue was
heated overnight at 105 ℃ and weighed. Lipid sam ̄
ples were analysed using a triple quadrupole tandem
mass spectrometry (MS / MS) equipped for electros ̄
pray ionisation (ESI). The lipids in each class were
quantified by comparison with two internal standards
for the class. Data processing was performed as pre ̄
viously described (Welti et al.ꎬ 2002). Five repli ̄
cates from each sampling time were analysed. A Q ̄
test was performed on the total amount of lipid in
each head ̄group classꎬ and data from discordant
samples were removed (Welti et al.ꎬ 2002). The
data were subjected to one ̄way analysis of variance
(ANOVA) using SPSS 16 0. Statistical significance
was tested by Fisher’ s least significant difference
(LSD) method. The double ̄bond index (DBI) was
calculated using the formula: DBI=(∑[N × mol%
lipid]) / 100ꎬ where N is the number of double bonds
in each lipid molecule (Rawyler et al.ꎬ 1999). Acyl
chain length (ACL) was calculated using a formula
derived from the DBI calculation above: ACL = (∑
[NC × mol% lipid]) / 100ꎬ where NC is the number
of carbon atoms in each lipid molecule.
The lipid contents of molecular species were
used for assessing the differences among the samples
(treatments and genotypes) by PCA with SPSS 16 0.
More than 120 lipid contents of molecular species
were transformed to a new set of variablesꎬ the prin ̄
cipal components were ordered so that the first few
retain most of the variation present in all of the origi ̄
nal variables (Jolliffeꎬ 2005).
2 Results
2 1 A thaliana altered its lipid content and com ̄
position in response to wounding
The wounded leaves of WS and PLDδ ̄KO were
sampled 0 and 30 minꎬ and 1ꎬ 3 and 6 h after woun ̄
ding. Lipids were profiled by a lipidomics approach
based on ESI ̄MS / MS. We identified quantitatively
more than 120 molecular species of glycerolipidsꎬ
which included six head ̄group classes of phospholip ̄
ids: phosphatidylcholine (PC)ꎬ phosphatidylethan ̄
olamine ( PE)ꎬ phosphatidylinositol ( PI)ꎬ phos ̄
phatidylserine ( PS)ꎬ PA and phosphatidylglycerol
(PG)ꎻ as well as two head ̄group classes of galacto ̄
lipids: monogalactosyldiacylglycerol ( MGDG) and
digalactosyldiacylglycerol (DGDG). Each molecular
species was identified in terms of the total number of
acyl carbon atoms and double bonds (Welti et al.ꎬ
2002). The total amount of lipid and the average
level of molecular species in each head ̄group class
are shown in Table 1 and Fig 1.
The lipid content changed significantly in the
leaves of Arabidopsis of both genotypes upon woun ̄
ding treatment. The levels of DGDGꎬ MGDGꎬ PC
and PI decreased significantly during the period after
wounding. The extraplastidic lipidsꎬ PGꎬ PCꎬ PE
and PIꎬ degraded much more in response to woun ̄
ding in PLDδ ̄KO than those in WS. It suggested
that the absence of PLDδ might enhance the degra ̄
dation of extraplastidic lipids in response to woun ̄
ding. The lipid class that showed the largest change
in absolute amount was PAꎬ which increased sharply
within 1 h after wounding treatmentꎬ and then gradu ̄
034 植 物 分 类 与 资 源 学 报 第 37卷
ally decreased in both genotypes. Its level was higher
in WS than in PLDδ ̄KO Arabidopsis leaves at 30 minꎬ
1 h and 3 h after wounding (Table 1). The level of
PA rose sharply by nine ̄ or thirteen ̄fold at 1 h in
WS and at 3 h in PLDδ ̄KO after woundingꎬ respec ̄
tively (Table 1). Profiling more than 120 molecular
species of membrane lipids in the leaves of both WS
and PLDδ ̄KO genotypesꎬ in terms of their absolute
contents (Fig 1)ꎬ revealed that the levels of molecu ̄
lar species 34∶6 ̄MGDG and 36∶6 ̄DGDGꎬ as well as
all detected molecular species of PCꎬ PG and PIꎬ
declined throughout the tested period after wounding.
Meanwhileꎬ levels of all PA molecular species in ̄
creased in both genotypes. Given that 34 ∶ 4 ̄PG was
the only molecular species detected with 34 ∶ 4 acyls
(Fig 1)ꎬ it was suggested that the increase in wound ̄
induced 34 ∶ 4 ̄PA was derived from PLD ̄mediated
conversion from PG. In factꎬ all of the molecular spe ̄
cies of PA that increased might have been derived
from PLD ̄mediated modification of PCꎬ PEꎬ PGꎬ PI
and PS (Wangꎬ 2005). Thereforeꎬ PA molecular spe ̄
cies appeared to reflect the rapid response of phospho ̄
lipids to wounding damage in Arabidopsis leaves.
2 2 Principal component analysis revealed dif ̄
ferent responses of lipid molecular species to woun ̄
ding
Differences among the samples ( treatments and
genotypes) were assessed using PCA with SPSS 16 0
The results of this analysis highlighted major statisti ̄
cally significant differences among the different peri ̄
ods after wounding. Analysis of the dataset allowed
the extraction of nine principal components that ex ̄
plained 100% of the variance in the system (Table
2). The first three componentsꎬ which are used for
plotting the scores and loadingsꎬ explained 75% of
the total variance (Table 2). Fig 2 depicts two score
plots.
Principal component 1 represents the lipid spe ̄
cies of PA which are the class exhibited the largest
changes and the lipid species of PC which are most
abundant constitute of extraplastidic membrane lipid
classes (Table 3)ꎻ principal component 2 represents
Table 1 Amount of lipid in each head ̄group class in WS and PLDδ ̄KO Arabidopsis leaves at different times after wounding. The percentage
of maximum relative change in lipids after wounding (Max RC) is the value for the maximum difference between the values of Control
and different period after woundingꎬ divided by the value of Control one. Values in the same row with different letters
are significantly different (P < 0 05) . Values are means ± standard deviation (n= 4 or 5)
Lipid
class Genotype
Lipid / dry weight (nmol / mg)
0 30 min 1 h 3 h 6 h
Max RC
/ %
DGDG WSPLDδ ̄KO
35 98 ± 1 24a
31 24 ± 1 14a
30 9 ± 2 92b
28 3 ± 3 46ab
30 42 ± 2 21bc
26 62 ± 2 39b
27 52 ± 0 73c
26 8 ± 4 44b
28 06 ± 2 09c
28 77 ± 1 7ab
-23 5
-14 8
MGDG WSPLDδ ̄KO
159 37 ± 3 53a
149 49 ± 10 3a
133 41 ± 15 3b
125 47 ± 13 24b
121 61 ± 28 39b
115 61 ± 11 48b
121 42 ± 11 75b
118 1 ± 14 85b
118 77 ± 8 71b
117 57 ± 11 41b
-23 8
-22 7
PG WSPLDδ ̄KO
11 2 ± 1 44a
11 5 ± 0 29a
10 51 ± 1 47a
9 49 ± 1 1b
10 61 ± 0 71a
7 7 ± 0 34c
9 98 ± 2 15a
8 64 ± 0 98bc
9 16 ± 1 64a
7 54 ± 1 87c
—
-34 4
PC WSPLDδ ̄KO
13 94 ± 1 62a
14 5 ± 1 72a
12 79 ± 1 94ab
12 19 ± 1 3b
12 03 ± 2 05ab
10 26 ± 0 14b
11 12 ± 3 72b
10 57 ± 0 43b
10 43 ± 0 8b
11 49 ± 1 25b
-25 2
-29 2
PE WSPLDδ ̄KO
4 37 ± 1 05a
5 53 ± 1 26a
5 11 ± 1 52a
5 37 ± 0 97ab
4 74 ± 1 21a
3 71 ± 0 44b
4 62 ± 1 51a
4 13 ± 0 78ab
4 4 ± 0 64a
4 57 ± 2 45ab
—
-32 9
PI WSPLDδ ̄KO
2 28 ± 0 28ab
2 35 ± 0 1a
2 47 ± 0 12ab
2 15 ± 0 31a
2 56 ± 0 48a
1 72 ± 0 28b
2 24 ± 0 06ab
2 04 ± 0 27ab
2 17 ± 0 16b
2 34 ± 0 23a
-4 8
-26 8
PS WSPLDδ ̄KO
0 27 ± 0 06a
0 29 ± 0 05ab
0 33 ± 0 09a
0 33 ± 0 05a
0 29 ± 0 06a
0 24 ± 0 05ab
0 29 ± 0 09a
0 23 ± 0 04b
0 24 ± 0 03a
0 25 ± 0 16ab
—
—
PA WSPLDδ ̄KO
0 4 ± 0 06c
0 23 ± 0 2d
2 98 ± 0 36b
2 31 ± 0 24c
4 17 ± 1 32a
2 85 ± 0 06b
4 11 ± 1 21a
3 31 ± 0 45a
2 68 ± 0 78b
2 94 ± 0 35ab
942 5
1339 1
1344期 Li Ai ̄hua et al.: Phospholipase Dδ is Involved in Wounding ̄Induced Phosphatidic Acid Formation in
Fig 1 Changes in the molecular species of membrane lipids in WS (A) and PLDδ ̄KO Arabidopsis (B)ꎬ
in terms of absolute contents. Values are means ± standard deviation (n= 4 or 5)
234 植 物 分 类 与 资 源 学 报 第 37卷
Table 2 The nine principal components extracted by
principal component analysis
Component
Extraction sums of squared loadings
Total % of Variance Cumulative %
1 51 837 42 144 42 144
2 24 702 20 083 62 227
3 15 789 12 836 75 064
4 8 884 7 223 82 286
5 6 721 5 464 87 750
6 5 155 4 191 91 942
7 4 237 3 445 95 387
8 3 517 2 859 98 246
9 2 158 1 754 100 000
the main plastidic component lipidsꎬ MGDG and DG ̄
DG (Fig 2)ꎻ principal component 3 basically repre ̄
sents lipids PE and PSꎬ which differentiate the two
genotypes (Fig 2). The highest and lowest loading
values (Table 3) are the lipid species that are most
important in the assignment of each principal compo ̄
nent. Examination of loadings (Table 3) clearly re ̄
veals that the separation of the points for the control
treatment and for the timing of 30 min from those for
1ꎬ 3 and 6 h along the principal component 1 results
primarily from the levels of PA and PC. There were
lower levels of PA species and higher levels of PC
species in fresh and 30 min post ̄wounding leaves
than those in leaves of both genotypes from other pe ̄
riods post ̄wounding. This indicates that these are the
most important lipid species for differentiation be ̄
tween the distinct periods after wounding. Principal
component 3 clearly describes differences between
the two genotypesꎬ WS and PLDδ ̄KO ( Fig 2B).
Examination of loadings (Table 3) reveals that the
separation between WS and PLDδ ̄KO along the
principal component 3 axis results primarily from the
higher levels of PS and PE species in PLDδ ̄KO and
the higher levels of 32∶1 ̄ and 34 ∶ 1 ̄PGꎬ and 34 ∶ 2 ̄
DGDG in WS (Fig 2)ꎬ which indicates that these
species are important in the statistical differentiation
of WS from PLDδ ̄KO.
Plot of principal component 1 against principal
component 2 showed the relationships and distances
between the two genotypes at different times after
wounding (Fig 2A). The control of WS and PLDδ ̄
KO cluster togetherꎬ and the points for 30 min post ̄
wounding of the two genotypes also cluster together.
Howeverꎬ the points for WS and PLDδ ̄KO at 1 h
and 3 h are scattered until at 6 h when they cluster
together again (Fig 2A). These patterns of separa ̄
tion and clustering basically occur along the princi ̄
pal component 1ꎬ which suggests that PA and PC
mainly contribute these patterns of variation. Along
principal component 3 ( Fig 2B)ꎬ the differences
between genotypes in same treatment were larger
than between treatments in same genotype. This sug ̄
gests that PE and PS do nothing with woundingꎬ but
contribute to the difference of two genotypes.
Fig 2 Principal component analysis (PCA) of lipid molecular species
in five time points post ̄wounding ( treatments called here) among WS
and PLDδ ̄KO Arabidopsis. The first three componentsꎬ which accoun ̄
ted for 75% of the total data varianceꎬ were used for plotting the scores.
Each pointꎬ a treatment and genotype combination is the mean of the
corresponding replicates’ principal component scores. (A) The score
plot of principal component 1 (42% of the variance) vs. principal com ̄
ponent 2 (20%). (B) The score plot of principal component 1 (42%
of the variance) vs. principal component 3 (13%). WSꎬ Wassilewski ̄
jaꎻ δꎬ PLDδ ̄KOꎻ 0ꎬ controlꎻ 30’ꎬ 30 minutes post ̄woundingꎻ 1 hꎬ 1 h
post ̄woundingꎻ 3 hꎬ 3 h post ̄woundingꎻ 6 hꎬ 6 h post ̄wounding
3344期 Li Ai ̄hua et al.: Phospholipase Dδ is Involved in Wounding ̄Induced Phosphatidic Acid Formation in
Table 3 Loadings of the first three principal components. The twelve highest and twelve lowest loading values are indicated
Lipid species Principal component 1loadings Lipid species
Principal component 2
loadings Lipid species
Principal component 3
loadings
Twelve lowest loading values
34 ∶ 6 PA -0 688 34 ∶ 5 MGDG -0 632 38 ∶ 5 PS -0 755
34 ∶ 0 PG -0 687 38 ∶ 5 MGDG -0 615 36 ∶ 6 PE -0 597
34 ∶ 4 PA -0 673 36 ∶ 5 DGDG -0 595 34 ∶ 1 PS -0 556
36 ∶ 1 DGDG -0 634 38 ∶ 4 PS -0 594 36 ∶ 4 PS -0 52
36 ∶ 2 PA -0 631 34 ∶ 6 DGDG -0 559 36 ∶ 5 PS -0 489
34 ∶ 3 PA -0 595 36 ∶ 5 MGDG -0 533 36 ∶ 1 PE -0 476
36 ∶ 3 PA -0 563 40 ∶ 3 PC -0 532 42 ∶ 2 PE -0 457
36 ∶ 5 PA -0 552 36 ∶ 6 DGDG -0 476 36 ∶ 2 PE -0 412
34 ∶ 2 PA -0 535 34 ∶ 4 DGDG -0 422 42 ∶ 4 PE -0 41
36 ∶ 6 PA -0 532 34 ∶ 4 MGDG -0 408 36 ∶ 5 PE -0 406
38 ∶ 6 DGDG -0 532 34 ∶ 6 MGDG -0 4 34 ∶ 1 PE -0 395
38 ∶ 6 MGDG -0 524 44 ∶ 2 PS -0 397 38 ∶ 2 PE -0 39
Twelve highest loading values
36 ∶ 5 PC 0 894 36 ∶ 2 DGDG 0 69 34 ∶ 6 PA 0 547
38 ∶ 2 PC 0 895 36 ∶ 4 PA 0 703 34 ∶ 3 MGDG 0 565
34 ∶ 5 DGDG 0 902 36 ∶ 1 MGDG 0 719 34 ∶ 1 MGDG 0 603
36 ∶ 4 DGDG 0 904 36 ∶ 5 PA 0 722 34 ∶ 4 PI 0 609
34 ∶ 4 PC 0 909 36 ∶ 6 PA 0 726 32 ∶ 0 PG 0 613
36 ∶ 3 PC 0 914 42 ∶ 3 PE 0 739 38 ∶ 2 PS 0 631
36 ∶ 6 MGDG 0 916 38 ∶ 6 MGDG 0 747 34 ∶ 3 PI 0 633
34 ∶ 3 DGDG 0 931 40 ∶ 3 PS 0 747 40 ∶ 2 PS 0 64
36 ∶ 4 PC 0 933 38 ∶ 3 MGDG 0 76 36 ∶ 6 PI 0 651
36 ∶ 2 PC 0 943 36 ∶ 2 MGDG 0 766 34 ∶ 1 PG 0 726
34 ∶ 3 PC 0 943 38 ∶ 4 MGDG 0 771 32 ∶ 1 PG 0 762
34 ∶ 2 PC 0 947 38 ∶ 6 PE 0 848 34 ∶ 2 MGDG 0 799
2 3 PA transiently increases and reacts differ ̄
ently to wounding in the WS and PLDδ ̄KO Ara ̄
bidopsis
The PCA result suggested that PA was one of the
main principal component factors involved in the re ̄
sponse to wounding in WS and PLDδ ̄KO Arabidopsis.
Thereforeꎬ we further used expression pattern analysis
with Mev 4 9 0 to analyse in details the variation of
PA species after wounding in the two genotypes.
The results show that the PA level transiently
increased upon wounding in the both genotypes
(Fig 3A-K). the levels of PA and several of its
main species in control leaves in PLDδ ̄KO were
similar to those of WS (Fig 3A-I). This suggests
that an absence of PLDδ does not influence the level
of PA in control leaves in Arabidopsis. At 30 min af ̄
ter woundingꎬ the content of PA and several of its
main species increased sharplyꎻ particularly large
increases were found for 36∶4 ̄ and 36∶5 ̄PAꎬ which
increased 30 ̄ and 17 ̄fold in PLDδ ̄KOꎬ and 27 ̄ and
17 ̄fold in WSꎬ respectively ( Fig 1Aꎬ B)ꎻ subse ̄
quentlyꎬ most of them continued to increase to the
highest level at 1 h after woundingꎬ and then de ̄
creased until 6 h after wounding (Fig 3A-I). How ̄
everꎬ there were still several differences of PA varia ̄
tion in response to wounding between the two geno ̄
types. Firstlyꎬ the levels of PA and nearly all of its
species in PLDδ ̄KO were lower than those in WS
(Fig 3A-Dꎬ F-I) from 30 min to 3 h after woun ̄
ding. In WSꎬ the PA level increased to the highest
level of 4 17 nmolmg-1ꎬ but that in PLDδ ̄KO was
3 31 nmolmg-1ꎬ this suggests that suppression of
PLDδ caused a 20% decrease of the PA level in Ara ̄
bidopsis. Secondlyꎬ the peak PA levels during the
post ̄wounding period differed: namelyꎬ the peak
occurred at 1 h in WSꎬ but at 3 h in PLDδ ̄KO
434 植 物 分 类 与 资 源 学 报 第 37卷
(Fig 3Jꎬ K). This suggested that a lack of PLDδ
slows the early response of PA. Finallyꎬ the differ ̄
ence of PA levels between the two genotypes to
wounding decreased over time. This suggested that
PLDδ had a rapid response to wounding.
2 4 The responses of unsaturation and acyl
chain length to wounding in WS and PLDδ ̄KO
Arabidopsis
The level of unsaturation level ( Quartacci et
al.ꎬ 2002) and acyl chain length (Denich et al.ꎬ
2003) of phospholipids and glycolipids in mem ̄
branes could affect membrane fluidity . The DBI was
calculated by the double bond of lipids and ACLs
was calculated by the total number of acyl carbon at ̄
oms of lipids. We analysed the DBI and ACL of lipid
classes in response to wounding in Arabidopsis (WS
and PLDδ ̄KO). After woundingꎬ the DBI of plastidic
membrane lipids PG decreased significantly in both
Fig 3 Changes of the levels of PA and its molecular species after wounding in WS and PLDδ ̄KO Arabidopsis leaves. A-I. Detected levels of
PA and its species (Aꎬ PAꎻ Bꎬ 34 ∶ 2 ̄PAꎻ Cꎬ 34 ∶ 3 ̄PAꎻ Dꎬ 34 ∶ 4 ̄PAꎻ Eꎬ 36 ∶ 2 ̄PAꎻ Fꎬ 36 ∶ 3 ̄PAꎻ Gꎬ 36 ∶ 4 ̄PAꎻ Hꎬ 36 ∶ 5 ̄PAꎻ
Iꎬ 36 ∶ 6 ̄PA). J-K. The curve upon combining eight PA species in WS (J) and PLDδ ̄KO (K). “∗” indicates that the value is
significantly different from that for WS under the same condition (P < 0 05) . Values are means ± standard deviation (n= 4 or 5)
5344期 Li Ai ̄hua et al.: Phospholipase Dδ is Involved in Wounding ̄Induced Phosphatidic Acid Formation in
genotypesꎬ whereas those of extra ̄plastidic mem ̄
brane lipids—PEꎬ PIꎬ PS and PA— almost in ̄
creased (Table 4). The maximum changing of DBI
after wounding was that of PGꎬ which decreased
0 08ꎬ and 0 11 in WS and PLDδ ̄KO separately.
The ACL of most lipid classes remained unchanged
after wounding. Howeverꎬ that of the plastidic lipid
(including MGDGꎬ DGDGꎬ and PG) changed sig ̄
nificantly after wounding (Table 5). Basicallyꎬ DBI
of lipid classes respond to woundingꎬ but only ACL
of plastidic lipids respond to wounding in Arabidopsis
(both WS and PLDδ ̄KO).
3 Discussion
After the profiling and analysis of lipids for a
relatively long period of 6 h after wounding in WS
and PLDδ ̄KO Arabidopsisꎬ we found that levels of
the majority of lipids were reduced rapidly in both
WS and PLDδ ̄KO leaves after wounding. Howeverꎬ
only level of PA increased significantly in response
to wounding. PLDδ also influenced the degradation
of some lipidsꎬ such as lipids PEꎬ PI and PGꎬ
which decreased their amounts sharply in PLDδ ̄KO
leavesꎬ but were unchanged at the same time points
in WS after wounding for later than 1 h. PLDδ partly
regulates PA levels induced by wounding. During the
whole period after woundingꎬ the most dramatic
changes of lipids occurred 30 min after wounding.
Thirty minutes is a limit timeꎬ but appears to be suf ̄
ficient for lipids in Arabidopsis to respond to wound
damageꎬ especially in the case of the lipid PA. This
rapid response suggests that PA might act as a sig ̄
nalling molecule in response to wounding. The rapid
lipid hydrolysis was the same as that reported previ ̄
ously in wounded tissues of various plants ( Lee et
al.ꎬ 1997ꎻ Ling et al.ꎬ 2007ꎻ Lojkowskaꎬ 1988ꎻ
Zien et al.ꎬ 2001). Howeverꎬ the trend of variation
of different lipid classes at the same time points dif ̄
feredꎻ for exampleꎬ in our work on Arabidopsisꎬ lip ̄
id PE did not change among the time points after
wounding in WSꎬ but it decreased significantly re ̄
ported in Zien et al. (2001).
Table 4 DBI values of membrane lipids in WS and PLDδ ̄KO Arabidopsis leaves at different times after wounding. The values
of maximum relative change (Max RC) is the maximum values for the difference between the DBI values after wounding
and control. Values in the same row with different letters are significantly different different (P < 0 05) .
Values are means ± standard deviation (n= 4 or 5)
Lipid
class Genotype
Double Bond Index (DBI)
0 30 min 1 h 3 h 6 h
Max RC
/ %
DGDG WSPLDδ ̄KO
5 43 ± 0 07a
5 37 ± 0 02a
5 35 ± 0 04b
5 37 ± 0 08a
5 39 ± 0 05a
5 4 ± 0 01a
5 38 ± 0 04a
5 4 ± 0 04a
5 39 ± 0 04a
5 34 ± 0 04a
-0 08
—
MGDG WSPLDδ ̄KO
5 94 ± 0 02a
5 95 ± 0 01b
5 95 ± 0 01a
5 96 ± 0 01a
5 95 ± 0 01a
5 96 ± 0 01a
5 94 ± 0 01a
5 95 ± 0 01ab
5 95 ± 0a
5 94 ± 0 01b
—
0 01
PG WSPLDδ ̄KO
3 43 ± 0 06a
3 45 ± 0 04a
3 37 ± 0 01a
3 38 ± 0 06b
3 36 ± 0 05a
3 34 ± 0 02b
3 35 ± 0 06b
3 35 ± 0 07b
3 35 ± 0 07b
3 41 ± 0 04ab
-0 08
-0 11
PC WSPLDδ ̄KO
2 99 ± 0 04a
3 03 ± 0 06a
2 97 ± 0 02a
3 03 ± 0 02a
3 ± 0 06a
3 02 ± 0 02a
3 02 ± 0 04a
3 06 ± 0 08a
3 03 ± 0 02a
3 02 ± 0 07a
—
—
PE WSPLDδ ̄KO
3 46 ± 0 01b
3 52 ± 0 07a
3 48 ± 0 03ab
3 51 ± 0 05a
3 48 ± 0 02ab
3 53 ± 0 03a
3 5 ± 0 01a
3 53 ± 0 03a
3 48 ± 0 03ab
3 55 ± 0 05a
0 04
—
PI WSPLDδ ̄KO
2 78 ± 0 02b
2 77 ± 0 03b
2 74 ± 0 03b
2 76 ± 0 04b
2 76 ± 0 03b
2 76 ± 0 02b
2 83 ± 0 02a
2 84 ± 0 05a
2 82 ± 0 04a
2 81 ± 0 07ab
0 05
0 07
PS WSPLDδ ̄KO
2 67 ± 0 02b
2 71 ± 0 05b
2 68 ± 0 05b
2 76 ± 0 03a
2 68 ± 0 05b
2 7 ± 0 02b
2 66 ± 0 02b
2 71 ± 0 04b
2 73 ± 0 02a
2 72 ± 0 02ab
0 06
0 05
PA WSPLDδ ̄KO
3 56 ± 0 38a
3 64 ± 0 08b
3 64 ± 0 05a
3 75 ± 0 06a
3 63 ± 0 15a
3 65 ± 0 05ab
3 67 ± 0 05a
3 66 ± 0 08ab
3 63 ± 0 06a
3 58 ± 0 12b
—
0 11
634 植 物 分 类 与 资 源 学 报 第 37卷
Table 5 ACL values of membrane lipids in WS and PLDδ ̄KO Arabidopsis leaves at different times after wounding. The values
of maximum relative change (Max RC) is the maximum values for the difference between the ACL values after wounding
and control. Values in the same row with different letters are significantly different (P < 0 05) .
Values are means ± standard deviation (n= 4 or 5)
Lipid
class Genotype
Acyl chain length
0 30 min 1 h 3 h 6 h
Max RC
/ %
DGDG WSPLDδ ̄KO
35 51 ± 0 04a
35 46 ± 0 02a
35 48 ± 0 03b
35 46 ± 0 02a
35 51 ± 0 04a
35 5 ± 0 04a
35 49 ± 0 02a
35 48 ± 0 04a
35 49 ± 0 03a
35 46 ± 0 03a
-0 03
—
MGDG WSPLDδ ̄KO
34 37 ± 0 01b
34 37 ± 0 03a
34 4 ± 0 02a
34 4 ± 0 02a
34 4 ± 0 02a
34 4 ± 0 03a
34 39 ± 0 03ab
34 37 ± 0 03a
34 39 ± 0 02ab
34 39 ± 0 05a
0 03
PG WSPLDδ ̄KO
33 89 ± 0 01a
33 92 ± 0a
33 87 ± 0 01b
33 87 ± 0 01b
33 86 ± 0 02b
33 87 ± 0b
33 86 ± 0 01b
33 86 ± 0 02b
33 85 ± 0 01b
33 89 ± 0 04ab
-0 04
-0 06
PC WSPLDδ ̄KO
35 29 ± 0 05a
35 33 ± 0 03a
35 29 ± 0 03a
35 28 ± 0 02b
35 29 ± 0 04a
35 31 ± 0 03ab
35 3 ± 0 04a
35 3 ± 0 06ab
35 3 ± 0 02a
35 29 ± 0 04ab
—
0 05
PE WSPLDδ ̄KO
35 25 ± 0 05a
35 29 ± 0 02a
35 29 ± 0 02a
35 33 ± 0 06a
35 29 ± 0 0a
35 3 ± 0 04a
35 25 ± 0 04a
35 28 ± 0 05a
35 26 ± 0 04a
35 33 ± 0 07a
—
—
PI WSPLDδ ̄KO
34 25 ± 0 06a
34 3 ± 0 03a
34 28 ± 0 02a
34 24 ± 0 08a
34 24 ± 0 04a
34 27 ± 0 03a
34 29 ± 0 04a
34 28 ± 0 03a
34 28 ± 0 05a
34 28 ± 0 02a
—
—
PS WSPLDδ ̄KO
38 29 ± 0 3a
38 32 ± 0 16a
38 08 ± 0 2a
38 13 ± 0 35a
38 38 ± 0 21a
38 39 ± 0 47a
38 34 ± 0 25a
38 4 ± 0 35a
38 36 ± 0 44a
37 76 ± 1 1a
—
—
PA WSPLDδ ̄KO
34 76 ± 0 12a
34 76 ± 0 23a
34 93 ± 0 04a
34 94 ± 0 11a
34 89 ± 0 06a
34 91 ± 0 08a
34 82 ± 0 04a
34 85 ± 0 03a
34 83 ± 0 07a
34 82 ± 0 07a
—
—
Among all the tested lipid classesꎬ PA exhibi ̄
ted the greatest increase in response to wounding.
Specificallyꎬ its level varied after woundingꎬ and the
patterns of variation were basically similar between
the two genotypes: a trend of a rise followed by a
fall. PLDα1 was another main member of the PLD
family that responded to wounding (Wang et al.ꎬ
2000). The level of PA still increased rapidly in
PLDδ ̄KOꎬ which suggested that the lack of PLDδ
did not influence the reaction of PLDα1 to woun ̄
ding. The function of PLDδ mainly exhibited during
the period of 30 min to 3 h after wounding (Fig 3Aꎬ
Table 1). It decreased the level of PA by about 20%
at 30 min after wounding in Arabidopsis ( Fig 3Aꎬ
Table 1)ꎬ which suggests that about 20% of PA pro ̄
duction at this time point in response to wounding is
attributable to PLDδ. Meanwhileꎬ Wang et al. (2000)
mentioned about 60% of wound ̄induced PA was at ̄
tributable to PLDα1ꎬ another member of the PLD
familyꎬ in Arabidopsisꎻ in that caseꎬ PA responded
to wounding five minutes after wounding treatment.
In summaryꎬ PLDδ has a role in wounding ̄induced
PA formation in Arabidopsisꎬ but it reacts later and
with less intensity than PLDα1.
Acknowledgement: The authors would like to thank Mary Roth
for the acquisition and processing of the ESI ̄MS / MS data.
Reference:
Bargmann BOꎬ Laxalt AMꎬ ter Riet B et al.ꎬ 2009. Multiple PLDs re ̄
quired for high salinity and water deficit tolerance in plants [J] .
Plant and Cell Physiologyꎬ 50 (1): 78—89
Denich Tꎬ Beaudette Lꎬ Lee H et al.ꎬ 2003. Effect of selected envi ̄
ronmental and physico ̄chemical factors on bacterial cytoplasmic
membranes [J] . Journal of Microbiological Methodsꎬ 52 ( 2):
149—182
Dhondt Sꎬ Geoffroy Pꎬ Stelmach BA et al.ꎬ 2000. Soluble phospho ̄
lipase A2 activity is induced before oxylipin accumulation in to ̄
bacco mosaic virus ̄infected tobacco leaves and is contributed by
patatin ̄like enzymes [ J] . The Plant Journalꎬ 23 (4): 431—
440
Hanahan DJꎬ Chaikoff Iꎬ 1947. A new phospholipide ̄splitting enzyme
specific for the ester linkage between the nitrogenous base and
the phosphoric acid grouping [J] . Journal of Biological Chemis ̄
tryꎬ 169 (3): 699—705
7344期 Li Ai ̄hua et al.: Phospholipase Dδ is Involved in Wounding ̄Induced Phosphatidic Acid Formation in
Jolliffe ITꎬ 2005. Pricipal Component Analysisꎬ 2nd edition [ M].
New York: Springer
Katagiri Tꎬ Takahashi Sꎬ Shinozaki Kꎬ 2001. Involvement of a novel
Arabidopsis phospholipase Dꎬ AtPLDδꎬ in dehydration ̄inducible
accumulation of phosphatidic acid in stress signalling [ J] . The
Plant Journalꎬ 26 (6): 595—605
Lee SMꎬ Suh Sꎬ Kim S et al.ꎬ 1997. Systemic elevation of phospha ̄
tidic acid and lysophospholipid levels in wounded plants [ J] .
The Plant Journalꎬ 12 (3): 547—556
Li Wꎬ Wang Rꎬ Li M et al.ꎬ 2008. Differential degradation of extra ̄
plastidic and plastidic lipids during freezing and post ̄freezing re ̄
covery in Arabidopsis thaliana [J] . Journal of Biological Chem ̄
istryꎬ 283 (1): 461—468
Ling Lꎬ Li Wꎬ Xu Zꎬ 2007. Prifiling the changes of lipid molecular
species in response to wounding in Arabidopsis thaliana (Crucife ̄
rae) by ESI ̄MS / MS [J] . Acta Botanica Yunnanica (云南植物
研究)ꎬ 29 (6): 671—676
Lojkowska Eꎬ 1988. Lipid ̄composition and post ̄wounding degradation
in potato slices from cultivars differing in susceptibility to autolysis
[J] . Potato Researchꎬ 31 (3): 541—549
Pappan Kꎬ Austin ̄Brown Sꎬ Chapman KD et al.ꎬ 1998. Substrate se ̄
lectivities and lipid modulation of plant phospholipase Dαꎬ  ̄βꎬ
and ̄γ [J] . Archives of Biochemistry and Biophysicsꎬ 353 (1):
131—140
Pappan Kꎬ Qin Wꎬ Dyer JH et al.ꎬ 1997a. Molecular cloning and
functional analysis of polyphosphoinositide ̄dependent phospho ̄
lipase Dꎬ PLDβꎬ from Arabidopsis [ J] . Journal of Biological
Chemistryꎬ 272 (11): 7055—7061
Pappan Kꎬ Zheng Sꎬ Wang Xꎬ 1997b. Identification and characteriza ̄
tion of a novel plant phospholipase D that requires polyphos ̄
phoinositides and submicromolar calcium for activity in Arabidopsis
[J]. Journal of Biological Chemistryꎬ 272 (11): 7048—7054
Qin Wꎬ Pappan Kꎬ Wang Xꎬ 1997. Molecular heterogeneity of phos ̄
pholipase D ( PLD) cloning of PLDγ and regulation of plant
PLDγꎬ  ̄βꎬ and ̄α by polyphosphoinositides and calcium [ J] .
Journal of Biological Chemistryꎬ 272 (45): 28267—28273
Quartacci MFꎬ Glišic Oꎬ Stevanovic B et al.ꎬ 2002. Plasma membrane
lipids in the resurrection plant Ramonda serbica following dehy ̄
dration and rehydration [J] . Journal of Experimental Botanyꎬ 53
(378): 2159—2166
Qu LJ (瞿礼嘉)ꎬ Gu HY (顾红雅)ꎬ Bai SN (白书农)ꎬ translation
( original author: Buchanan BBꎬ Gruissem Wꎬ Jones RL)ꎬ
2004. Biochemistry & Molecular Biology of Plant (植物生物化
学与分子生物学)ꎬ Beijing: Science Press
Rawyler Aꎬ Pavelic Dꎬ Gianinazzi C et al.ꎬ 1999. Membrane lipid in ̄
tegrity relies on a threshold of ATP production rate in potato cell
cultures submitted to anoxia [ J] . Plant Physiologyꎬ 120 (1):
293—300
Ryu SBꎬ Wang XMꎬ 1996. Activation of phospholipase D and the pos ̄
sible mechanism of activation in wound ̄induced lipid hydrolysis
in castor bean leaves [ J ] . Biochimica et Biophysica Acta
(BBA)  ̄Lipids and Lipid Metabolismꎬ 1303 (3): 243—250
Wang Cꎬ Zien CAꎬ Afitlhile M et al.ꎬ 2000. Involvement of phospho ̄
lipase D in wound ̄induced accumulation of jasmonic acid in Ar ̄
abidopsis [J] . The Plant Cellꎬ 12 (11): 2237—2246
Wang Xꎬ 1999. The role of phospholipase D in signaling cascades
[J] . Plant Physiologyꎬ 120 (3): 645—652
Wang Xꎬ 2000. Multiple forms of phospholipase D in plants: the gene
familyꎬ catalytic and regulatory propertiesꎬ and cellular func ̄
tions [J] . Progress in Lipid Researchꎬ 39 (2): 109—149
Wang Xꎬ 2004. Lipid signaling [J] . Current Opinion in Plant Biolo ̄
gyꎬ 7 (3): 329—336
Wang Xꎬ 2005. Regulatory functions of phospholipase D and phospha ̄
tidic acid in plant growthꎬ developmentꎬ and stress responses
[J] . Plant Physiologyꎬ 139 (2): 566—573
Wang Xꎬ Xu Lꎬ Zheng Lꎬ 1994. Cloning and expression of phosphati ̄
dylcholine ̄hydrolyzing phospholipase D from Ricinus communis
L [J] . Journal of Biological Chemistryꎬ 269 (32): 20312—
20317
Wasternack Cꎬ Stenzel Iꎬ Hause B et al.ꎬ 2006. The wound response
in tomato ̄role of jasmonic acid [ J] . Journal of Plant Physiolo ̄
gyꎬ 163 (3): 297—306
Welti Rꎬ Li Wꎬ Li M et al.ꎬ 2002. Profiling membrane lipids in plant
stress responses ̄Role of phospholipase Dα in freezing ̄induced
lipid changes in Arabidopsis [ J] . Journal of Biological Chemis ̄
tryꎬ 277 (35): 31994—32002
Xu Lꎬ Zheng Sꎬ Zheng L et al.ꎬ 1997. Promoter analysis and expres ̄
sion of a phospholipase D gene from castor bean [ J] . Plant
Physiologyꎬ 115 (2): 387—395
Zhang Wꎬ Wang Cꎬ Qin C et al.ꎬ 2003. The oleate ̄stimulated phos ̄
pholipase Dꎬ PLDδꎬ and phosphatidic acid decrease H2 O2  ̄in ̄
duced cell death in Arabidopsis [J] . The Plant Cell Onlineꎬ 15
(10): 2285—2295
Zhang Wꎬ Yu Lꎬ Zhang Y et al.ꎬ 2005. Phospholipase D in the signa ̄
ling networks of plant response to abscisic acid and reactive oxy ̄
gen species [J] . Biochimica et Biophysica Acta (BBA)  ̄Molecu ̄
lar and Cell Biology of Lipidsꎬ 1736 (1): 1—9
Zien CAꎬ Wang Cꎬ Wang X et al.ꎬ 2001. In vivo substrates and the
contribution of the common phospholipase Dꎬ PLDαꎬ to wound ̄
induced metabolism of lipids in Arabidopsis [ J] . Biochimica et
Biophysica Acta ( BBA)  ̄Molecular and Cell Biology of Lipidsꎬ
1530 (2 ̄3): 236—248
834 植 物 分 类 与 资 源 学 报 第 37卷