| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
First published online February 24, 2002; 10.1104/pp.010928
Plant Physiol, March 2002, Vol. 128, pp. 1057-1068
The Arabidopsis Phospholipase D Family. Characterization of a
Calcium-Independent and Phosphatidylcholine-Selective PLD
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
ABSTRACT |
|---|
 TOP
 ABSTRACT
 INTRODUCTION
RESULTS AND DISCUSSION
MATERIALS AND METHODS
LITERATURE CITED
|
|---|
Four types of phospholipase D (PLD), PLD
, 
, 
, and 
,
have been characterized in Arabidopsis, and they display different requirements for Ca2+, phosphatidylinositol
4,5-bisphosphate (PIP2), substrate vesicle composition,
and/or free fatty acids. However, all previously cloned plant PLDs
contain a Ca2+-dependent phospholipid-binding C2 domain and
require Ca2+ for activity. This study documents a new type
of PLD, PLD
1, which is distinctively different from previously
characterized PLDs. It contains at the N terminus a Phox homology
domain and a pleckstrin homology domain, but not the C2 domain. A
full-length cDNA for Arabidopsis PLD1 has been identified and used
to express catalytically active PLD in Escherichia coli.
PLD1 does not require Ca2+ or any other divalent cation
for activity. In addition, it selectively hydrolyzes
phosphatidylcholine, whereas the other Arabidopsis PLDs use several
phospholipids as substrates. PLD1 requires PIP2 for
activity, but unlike the PIP2-requiring PLD or ,
phosphatidylethanolamine is not needed in substrate vesicles. These
differences are described, together with a genomic analysis of 12 putative Arabidopsis PLD genes that are grouped into , , ,
, and based on their gene architectures, sequence similarities,
domain structures, and biochemical properties.
| |
INTRODUCTION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
RESULTS AND DISCUSSION
MATERIALS AND METHODS
LITERATURE CITED
|
|---|
Phospholipase D (PLD) is a prevalent
family of phospholipases in plant tissues, and it cleaves
phospholipids, producing phosphatidic acid and a free head group such
as choline (Wang, 2000
). PLD was first identified in plants more than
50 years ago (Hanahan and Chaikoff, 1947). It did not receive
widespread attention in other organisms until the 1980s when PLD was
revealed to be activated by external stimuli and was later recognized
as a lipid-signaling enzyme together with phospholipase
A2, phospholipase C, and sphingomyelinase (Cockcroft, 1984; Bocckino et al., 1987). Since the first cloning of a
PLD cDNA from castor bean (Ricinus communis; Wang et al., 1994), understanding of PLDs at the molecular, biochemical, and cellular levels has since advanced greatly in plants, animals, and
fungi (Frohman et al., 1999; Liscovitch et al., 2000; Wang, 2000;
Munnik, 2001). PLD has been proposed to play a pivotal role in many
cellular processes such as signal transduction, membrane trafficking,
cytoskeletal rearrangements, and membrane degradation.
Multiple PLDs have been identified in plants. Four types of Arabidopsis
PLDs, PLD, , 1, and , have been characterized at the
molecular biological and biochemical levels (Pappan et al., 1997a,
1997b, 1998; Qin et al., 1997; Wang and Wang, 2001). PLD represents
the conventional plant PLD, which does not require phosphoinositides
for activity when assayed at millimolar levels of
Ca2+ (Pappan and Wang, 1999). In contrast, PLD
and 1 are phosphatidylinositol 4,5-bisphosphate
(PIP2) dependent and have maximum activity at micromolar levels of Ca2+ (Pappan et al., 1997a;
Qin et al., 1997). Recently identified PLD displays unique
biochemical properties; it is activated by free oleic acid and is
tightly associated with the plasma membrane (Wang and Wang, 2001).
Despite many differences in the biochemical properties, all of the
previously cloned plant PLDs contain a
Ca2+-dependent phospholipid-binding C2 domain
(protein kinase C-conserved 2 domain) and require
Ca2+ for activity (Wang, 2000). In addition, they
have broad substrate specificity, hydrolyzing several common membrane
phospholipids, phosphatidylcholine (PC), phosphatidylethanolamine (PE),
phosphatidylglycerol (PG), and phosphatidyl-Ser (PS; Pappan et al.,
1998). This study has identified a new type of PLD, PLD1, which is
independent of Ca2+ and selectively hydrolyzes
PC. PLD1 contains a Phox homology (PX) domain and a pleckstrin
homology (PH) domain, but not the C2 domain. The different biochemical
properties and domain structures are described, together with a genomic
analysis of the evolutionary relationships, sequence similarities, and
gene architectures of 12 PLD genes in Arabidopsis.
| |
RESULTS AND DISCUSSION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
RESULTS AND DISCUSSION
MATERIALS AND METHODS
LITERATURE CITED
|
|---|
Genomic Organization and Grouping of the Arabidopsis PLD Gene Family
Twelve PLD genes were identified in the Arabidopsis genome by
BLAST searching against the cloned cDNAs. Two of these genes are
located on chromosome I, one on chromosome II, three on
chromosome III, five on chromosome IV,
and one on chromosome V (Fig. 1; Table I). They are grouped
into five classes, PLD(1, 2, 3, 4), (1, 2), (1, 2, 3), ,
and (1, 2), based on the
gene architectures (Fig. 1), sequence similarities (Fig. 2; Table
II), domain structures (Figs. 3 and
4), and biochemical properties (Table
III). This classification updates an
earlier one (Wang, 2000) in light of the new sequence information and
analysis. In particular, the previously designated 1 has
been regrouped into the class as 2, and
the previous 2 is not included at this time, as its
putative genomic sequence is not annotated in the GenBank. The current
PLD encodes a newly characterized PLD that exhibits
unique properties (Katagiri et al., 2001; Wang and Wang, 2001).
|
|
|
|
|
|
|
Of the four genes, PLD1 and 2 are very
similar in terms of gene structure and sequence similarity. Their
deduced amino acid sequences share about 92% similarity (Table II).
Comparison of the cloned PLD cDNAs of castor bean and Arabidopsis
PLD1 with their genomic sequences has shown the presence
of an intron in the 5'-untranslated region of PLD1 (Xu et
al., 1997). Arabidopsis EST database searches against these two genes
revealed a number of EST clones corresponding to 1, but
none for 2 (Table I), indicating that they might differ
greatly in abundance despite their high degree of sequence similarity.
The complete cDNA for PLD1 has been isolated (Dyer et
al., 1995), and its nucleotide and deduced amino acid sequences show
some discrepancies from those based on the genomic exons. Verification
of this and other PLD cDNA sequences is under way to determine which
differences result from polymorphism or sequencing inaccuracies.
The PLD3 and 4 genes contain four exons in
their coding region (Fig. 1). PLD3 shares about 70% protein
sequence similarity to 1 and 2. However, to date, no EST clone is
yet identified for this gene. The deduced amino acid sequence of
PLD4 is shorter than PLD1, 2, and 3 (762 versus 810 or 820 amino
acids). It shows no more than 55% similarity to the other three s,
or to any other member of Arabidopsis PLDs. EST clones for
PLD4 are present in the database, so it is expressed.
PLD3 and 4 are grouped into the class
because they have four exons, whereas other PLDs have 10 or
more exons (Fig. 1). In addition, their overall deduced amino acid
sequences are evolutionarily closer to those of PLD1and 2 than to other PLDs (Fig. 2). Furthermore, their domain
structures involved in catalysis (Fig. 4), Ca2+
binding (Fig. 5), and
PIP2 interaction (Fig.
6) are more closely related to those of
PLD1 and 2 than to the other PLDs. Thus, they may have similar biochemical properties to PLD. However, the
amino acid sequence of PLD4 is quite distant from all PLDs (Table
II), and it might possess unique regulatory and catalytic properties.
Therefore, its designation to the PLD group is tentative, and its appropriate position in the Arabidopsis PLD gene family awaits
further experimental characterization of its domain structures and
biochemical properties.
|
|
PLD, , and genes all consist
of 10 exons in the coding region. The previously cloned and
characterized PLD is now designated PLD1,
and the amino acid sequences of PLD1 and 2
are about 89% similar (Table II). The annotated gene structure of
2 in the GenBank contains 11 exons, which is why it was
previously designated as a new class (Wang, 2000). However, a close
examination of the GenBank annotation indicates that the original
prediction for the presence of the first intron is somewhat ambiguous.
Inclusion of this intron removes part of an open reading frame that
encodes the beginning part of the PLD C2 domain. Changing this region to be part of the first exon would increase the overall similarity between 2 and 1 from 85% to 89%. Thus, the first intron is
removed in Figure 1, although ultimate determination of the exon-intron structures of PLD2 awaits the cloning of its cDNA. EST
database searching against the 2 gene has not identified
any expressed sequence. However, northern blotting probed with a
2-specific fragment showed a weak but positive result (C. Qin and X. Wang, unpublished data), suggesting at least that the
2 gene is expressed in low abundance under certain circumstances.
The three PLD genes cluster in tandem on chromosome IV
with a similar pattern of exon-intron spacing. The three
PLD genes are very closely related, sharing almost 95%
sequence similarities in the deduced amino acid level (Table II). The
PLD gene whose cDNA was first cloned and characterized
was designated 1. A complete cDNA for the second
PLD gene, designated 2, was cloned
recently, and some of its biochemical properties are the same as those
of 1 (C. Qin and X. Wang, unpublished data). The third
gene in the PLD gene cluster is designated as PLD3.
EST clones corresponding to the PLD sequences have been
identified, indicating that they all are expressed.
PLD is a single gene class (Fig. 1). The gene annotated
in the GenBank is actually much longer, with six more exons on the 3'
end that encodes a nuclear localization signal and an RNA-binding region. However, experimental evidence from cDNA cloning and transcript size has shown that the authentic PLD gene is shorter
than what was annotated and that does not contain the six extra exons
at the 3' end (Wang and Wang, 2001). Two full-length cDNAs of PLD have been cloned recently; they differ by 33 nucleotides near the end
of the second exon (Katagiri et al., 2001; Wang and Wang, 2001). The
cDNA that is 33 nucleotides shorter encodes a catalytically active PLD
(Wang and Wang, 2001). It is possible that the two cDNAs come from two
transcripts resulting from alternative splicing of the same PLD
gene. The names PLDa and PLDb are proposed to distinguish the two
PLD transcripts and proteins (Wang and Wang, 2001). The alphabetic
affixes distinguish different products of the same gene, whereas the
Arabic numerals represent separate PLD genes. A 90-kD Arabidopsis PLD
(accession no. AF306345) has recently been shown to be associated with
microtubules and has been implicated in regulating cell division
(Gardiner et al., 2001). Its cDNA and deduced amino acid sequences
exactly match those of PLDb (accession no. AF322228).
The last two members of putative PLD genes in the Arabidopsis genome
are grouped together as PLD1 and 2 (Fig. 1)
because of their similar protein domain structures (Fig. 3B). Both
genes are located on chromosome III. The two PLDs share about 74%
sequence similarity in deduced amino acids, and both are quite distant to all the other Arabidopsis PLDs, with no more than 46% sequence similarity (Table II). In addition, their deduced amino acid sequences are 1,096 and 1,039 amino acids in length, respectively, which are
longer than the other plant PLD proteins, but in the same range as the
mammalian PLD1 (human PLD1 is 1,074 amino acids). Sequence comparison
also suggests that PLDs are closer to the mammalian PLDs than to the
other plant ones in terms of amino acid sequence similarity (Table II)
and the protein domain structures (Figs. 3-5). EST clones have been
identified for both PLD genes.
Domain Structures of the Arabidopsis PLD Proteins
HKD Motif
The highly conserved domain in the PLD family is the HKD motif, which is used to define the PLD superfamily. It was termed "HKD" because the domain contains the motif HxKxxxxD/E, which is found twice without exception in all cloned PLDs (Hammond et al., 1995; Koonin,
1996; Ponting and Kerr, 1996). The two HKD motifs are far from each
other in the primary structure (Fig. 3), but they interact
with each other to form the active site. Point mutagenesis of PLD from
several species has revealed that these amino acids are critical for
catalysis in vitro and for PLD function in vivo (Sung et al., 1997;
Gottlin et al., 1998). This has led to a model for the catalytic cycle
using the conserved His for nucleophilic attack on the substrate
phosphorous and involving a covalent phosphatidyl enzyme intermediate.
The crystal structure of the bacterial endonuclease member Nuc of the
PLD superfamily, which encodes a single and very divergent HKD motif,
has been determined (Stuckey and Dixon, 1999). Recent structural
studies have also led to expansion of the second HKD motif of the PLD family to HxKxxxxDxxxxxxGSxN. However, definition of the precise role
that each amino acid plays will await determination from structural
analyses of an eukaryotic PLD crystallized with its normal substrate.
These duplicated HKD motifs are conserved in deduced amino acid
sequences of all the 12 Arabidopsis PLD genes. They are separated by
approximately 320 amino acids, except for 1 and 2, which have
longer spacing sequences (Fig. 3C). Sequence alignment reveals that
although all the second HKD motif structures look highly identical, the
first HKDs and flanking regions display more diversity (Fig. 4A). The
10 exon-containing PLDs, four exon-containing PLDs, and PLDs form
three distinct groups (Fig. 4B). The first HKD domains in the group
are more similar to that of hPLD1, again indicating a closer
relationship between PLD genes and the mammalian PLDs. The
dendrogram of the clustering relationship of the HKD1 domains (Fig. 4B)
is consistent with the phylogenetic tree of the whole proteins (Fig.
2).
"IYIENQFF" Motif
The "IYIENQFF" region between the two HKD motifs is also called conserved region III in mammalian PLD structures and is found only in family members that exhibit bona fide PLD activity (Frohman et al., 1999). Mutagenesis studies have demonstrated that it is almost as
critical as the HKD domains (Sung et al., 1997), but the precise role
has yet to be determined. This region may be responsible for
hydrophobic interactions with the methyl groups of the choline head
group, due to the enrichment of aromatic amino acids. An alternative
hypothesis is that it might encode the targeting signal for caveolae
(Frohman et al., 1999). This "IYIENQFF" motif is the most conserved
domain with the least diversity among the 12 Arabidopsis PLDs. This
sequence in the PLD group is exactly the same as that in mammalian
PLDs. For the other PLDs, only the seventh residue, Phe(F), is
substituted by a Tyr(Y), which should not alter the property of this region.
Ca2+/Phospholipid-Binding C2 Domain
All previously cloned plant PLDs have a C2 domain near the N terminus, which is unique to plant PLDs and is not present in animal or fungal PLDs. The C2 domains are approximately 130 residues in length and can bind Ca2+ and other effectors, including phospholipids, inositol phosphates, and proteins (Nalefski and Falke, 1996; Rizo and Sudhof, 1998; Zheng et al., 2000). The three-dimensional
structures and the Ca2+-binding properties of C2
domains in synatagmin (Shao et al., 1996), cPLA2 (Perisic et al., 1998;
Xu et al., 1998), PLC1 (Essen et al., 1997), and PKC (Sutton and
Sprang, 1998) have been well studied by x-ray crystallography and/or
NMR. Although the C2 domains have two topological variants, their
three-dimensional structures are strikingly conserved.
Ca2+ binding is coordinated by four to five amino
acid residues in bipartite loops within the C2 domain.
Arabidopsis PLDs, s, and all have the
Ca2+-coordinating acidic residues, whereas the
PLD C2 domains lack one or more of these potential
Ca2+ ligands (Fig. 5). Compared with
cPLA2, PLD1 has one, and 2 and 3 have
two of them substituted, whereas 4 contains none of the
Ca2+-binding residues. The presence of C2 domain
in PLD4 is not clear, but several reserved hydrophobic amino acids,
which have been proposed to maintain the structural integrity of the C2
fold, are present in its corresponding region. On the other hand,
Conserved Domain Database (CDD) searching and Gap comparison did not
reveal the existence of the C2 domain in the two PLDs.
The differences in the Ca2+-binding residues
suggests that Ca2+ affinity of PLDs could be
lower than that of , , and . This lower
Ca2+ affinity of PLD-C2 than PLD-C2 domain
has been demonstrated experimentally (Zheng et al., 2000).
Ca2+- and PC-binding properties of PLD C2 and
PLD C2 follow a trend similar to the Ca2+
requirements of the whole enzymes, PLD1 and PLD1, for PC
hydrolysis (Table III). Thus, it has been suggested that the C2 domains
of PLD and PLD serve as handles by which
Ca2+ differentially regulates these PLD
activities (Zheng et al., 2000).
PIP2-Binding Motifs
PIP2 plays a critical role in PLD activation in mammals and plants. PIP2 is required for activities of Arabidopsis PLD and ; replacement of
PIP2 by other phospholipids such as PC, PS, PG,
PE, and phosphatidylinositol (PI) resulted in loss of PLD activity (Qin
et al., 1997). Two putative PIP2-binding motifs were also predicted, flanking the second HKD domain of PLD (Pappan et al., 1997b), but the PIP2-binding ability of
these motifs in PLD has not been verified experimentally. Recent
studies on mouse PLD2 have suggested that a small region containing
many conserved basic and hydrophobic residues located between the two
HKD motifs is responsible for the PIP2
requirement (Fig. 6; Sciorra et al., 1999). Mutagenesis experiments
revealed that this region is required for activation of mammalian and
yeast PLDs by PIP2.
Sequence alignments of PLDs from plants, animals, and yeast reveals
conservation and variation of the basic residues of this PIP2-binding motif in plant PLDs (Fig. 6). The
clustering relationship is quite similar to that revealed by comparison
of HKD1 domains. For the five conserved basic amino acids originally
identified in mouse PLD2, human PLD1, and yeast PLD (Sciorra et al.,
1999), only the last one is substituted by an acidic residue in the two PLDs (Fig. 6). PLDs and s all conserved the last three
residues, but the first two are replaced by negatively charged residues or by neutral ones. PLD retains two of the conserved residues, whereas PLDs have only one of them (Fig. 6). A recent study using PLD1 has provided experimental evidence for this motif as a
PIP2-binding region, and this binding is
stimulated by Ca2+ (L. Zheng, R. Krishnamoorthi,
and X. Wang, unpublished data). It should be noted that the N-terminal
C2 domain of PLDs also binds PIP2, but unlike the
PIP2-binding region, the
C2-PIP2 interaction is inhibited by
Ca2+ (Zheng et al., 2000).
PX Domain
A conserved PX domain is present in Arabidopsis PLD1 and 2,
as well as in mammalian PLDs, but is not found in the other Arabidopsis
PLDs (Fig. 3B). The PX domain was originally identified in the protein
p47phox, a component of phagocyte NADPH oxidase
(Ponting, 1996). It is a novel protein module containing a conserved
Pro-rich motif. Recent structural and cell biological studies show that
the PX domain can bind phosphoinositides and SH3 domain (Cheever et
al., 2001; Hiroaki et al., 2001; Kanai et al., 2001). Therefore, it may
play a critical role in coordinating membrane localization and protein
complex assembly during cell signaling. The PX domain is critical for
activities of mammalian PLDs (Sung et al., 1999a, 1999b). How it
functions is unknown, although potential roles include regulating
interactions with factors that promote translocation or activation.
PH Domain
PH domain is another potential regulatory module present in Arabidopsis PLD1 and 2, as well as in mammalian PLDs, but not in
any other Arabidopsis PLDs (Fig. 3B). PH domains are composed of
approximately 120 amino acids found in more than 100 proteins involved
in cell signaling, cytoskeletal rearrangement, and other processes
(Lemmon and Ferguson, 2000). All PH domains studied to date appear to
bind to phosphoinositides, but most bind weakly and nonspecifically.
Only a small subclass of PH domains show strong and specific binding to
membrane phosphoinositides (Lemmon and Ferguson, 2000). Deletion
analysis of mammalian PLD1 and PLD2 revealed that the PH domain is not
required for enzymatic activity nor is the dependence of the enzymatic
activity on PIP2 affected by deletion of the PH
domain (Sung et al., 1999a, 1999b). PH domain could be important in
membrane targeting and association with other cellular components.
However, its function in PLDs awaits further investigation.
PLD1 Is a Calcium-Independent, PC-Selective PLD
To characterize the PX/PH-containing, putative PLDs, the
Arabidopsis EST database was searched for putative full-length cDNA clones of PLDs using the PLD coding sequences
annotated in the GenBank database. Several putative PLD1 EST clones
were identified, and one (EST no. AV529766) was sequenced completely
and found to be a full-length PLD1 cDNA. The cDNA is composed of
3,785 nucleotides, and the nucleotide sequence of the cDNA matches that annotated from the genomic sequencing. The coding region of the PLD1
cDNA starts at nucleotide 248 and ends at 3,538 (GenBank accession no.
AF411833). It encodes a protein of 1,096 amino acids with the domain
structures depicted in Figure 3B. The calculated molecular mass and pI
are 124 kD and 6.27, respectively.
To validate that this cDNA encodes a PLD, protein from the cDNA was
expressed in Escherichia coli using pBluescript SK(-) as
expression vector, which has been used successfully to express catalytically active PLD, , and 1 (Wang et al., 1994; Pappan et al., 1997b; Qin et al., 1997). After
isopropyl-1-thio--D-galactopyranoside induction, protein extracts from E. coli JM109 harboring the
SK alone exhibited negligible PLD activity (Fig.
7A). Proteins from E. coli
harboring the vector with the PLD1 cDNA insert had significant PLD
activity. PLD1 required PIP2 for activity; it
displayed no activity when PC-only vesicles and PC/Triton X-100
vesicles were used as substrates (Fig. 7A). The
PIP2 requirement is a property shared by PLD
and , as well as mammalian and yeast PLDs, but the activity of
PLD and also needs PE (Table III; Pappan et al., 1998). In
contrast, PLD1 requires only PIP2, but not PE (Fig. 7A).
|
Previous analysis of PLD and PLD indicates that the C2 domain is
a major determinant for the requirement of different levels of
Ca2+ for their activities (Zheng et al., 2000).
Accordingly, the absence of the C2 domain in PLD1 (Fig. 3B) could
mean that this PLD might not require Ca2+ for
activity. To test this hypothesis, 2 mM EGTA and 2 mM EDTA were added to the reaction to chelate
Ca2+ and any other divalent cation. In addition,
varied concentrations of CaCl2 or
MgCl2 were used to examine the cation effect on
PLD1. The highest PLD1 activity occurred in the zero to nanomolar
concentrations of Ca2+ and
Mg2+ (Fig. 7B). The activity decreased gradually
at the micromolar range and dropped rapidly when cation concentrations
approached millimolar levels (Fig. 7B). These results show that PLD1
is independent of Ca2+ or any other cation for
activity. This property is in contrast to all the other cloned plant
PLDs that require micromolar or millimolar ranges of
Ca2+ for activity (Wang, 2000; Table III).
The substrate specificity was examined next. PLD1 hydrolyzed PC
well, but had negligible activity toward PE or PS (Fig.
8A). No PG-hydrolyzing activity was
observed (data not shown). This PC-selective activity is distinctively
different from other characterized Arabidopsis PLDs, but similar to the
cloned mammalian ones. PLD hydrolyzes PC, PE, and PG equally well,
and PLD and hydrolyze PC, PE, and PS (Pappan et al., 1998; Table
III), whereas PLD uses PE better than PC (C. Qin, C. Wang, and X. Wang, unpublished data). The mammalian PLD1 is PC specific and has no
activity toward PE or PI (Hammond et al., 1995), and activity of the
cloned mammalian PLD2 toward lipids other than PC has not been
reported.
|
PLD1 functioned at a rather narrow range of pH with a pH optimum at
7. Its activity decreased considerably at pH 6.5 and 7.5, and was
virtually abolished at pH 6 and 8 (Fig. 8B). PLD and 1 are also
most active at pH 7, but their functional pH is broader than that of
PLD1 (Pappan and Wang, 1999; Table III). PLD has a comparable
activity from pH 5.5 to 8.5. PLD has a similar activity at pH 6.5 to
7.5, and has still approximately 35% activity at pH 5. PLD has
acidic pH optima that are influenced by the Ca2+
concentrations. The pH optimum is 4.5 to 5 when assayed in the presence
of micromolar levels of Ca2+ and
PIP2, but it increases to 5.5 to 6.5 at
millimolar levels of Ca2+ (Pappan and Wang,
1999). However, the presence or absence of Ca2+
did not alter the pH optimum of PLD1 (Fig. 8B).
In summary, this study has provided molecular and biochemical evidence
for the occurrence of a new class of plant PLD, PLD, that is
Ca2+ independent and PC selective. These results,
together with the genomic analysis of the Arabidopsis PLD family,
indicate that different PLDs are subjected to unique controls and that
their activation may hydrolyze different membrane lipids in the cell. Also noteworthy is that the biochemical properties, domain structures, and genomic organization of plant PLDs are much more diverse than for
other organisms. Thus far, only two PLD genes, PLD1 and
PLD2, are identified in mammals, and one PLD gene,
SPO14, has been identified in yeast. This is in contrast to
the other phospholipase families; more diversity in structures and
regulatory properties has been observed for PI-PLC and
PLA2 in animals than in plants (Wang, 2001).
Could this mean that plants use PLD more than other organisms as part
of the regulatory machinery in cellular functions? The occurrence of
the C2, PH, and PX domains, which have been demonstrated in many
proteins involved in signal transduction, vesicular trafficking, and
cytoskeletal rearrangements, hints at important and diverse roles of
PLDs in cellular regulation. Studies involving genetic manipulation of
specific PLDs have revealed some unique metabolic and physiological
function for some PLDs (Wang et al., 2000; Wang, 2001). Genetic
redundancy may also occur with some of the PLD genes (Katagiri et al.,
2001). Delineation of the unique functions and potential redundancy of
the multiple PLDs will be important to a comprehensive understanding of
the PLD family and their roles in plant processes.
| |
MATERIALS AND METHODS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
RESULTS AND DISCUSSION
MATERIALS AND METHODS
LITERATURE CITED
|
|---|
Identification of PLD and Sequence Analysis
A putative Arabidopsis EST PLD1 cDNA was identified by
searching the BLAST database against the PLD1 cDNA annotated from its genomic DNA sequence. This EST clone was kindly provided by the
Kazusa DNA Research Institute (Chiba, Japan). Gene-specific primers and
the T3 primer from the pBluescript SK(-) vector were used for the
sequencing of the insert, which was proved to be a full-length PLD1
cDNA (accession no. AF411833).
The PLD gene sequences were obtained by BLAST searching in GenBank and The Institute for Genomic Research databases using identified PLD cDNAs. This phylogenetic tree was generated using the GrowTree program from the University of Wisconsin Genetics Computer Group. The dendrogram was generated by the PileUp program in GCG. Domain structures were determined using CDD searching and Gap comparison.
Expression of PLD cDNA in Escherichia coli
Expression of the PLD1 cDNA was performed using pBluescript
SK(-) containing the cDNA insert in E. coli. The
recombinant plasmid was transformed into E. coli JM109.
Fifty microliters of an overnight culture of the transformed E.
coli was added to 25 mL of Luria Bertani medium with 50 µg
mL
1 ampicillin. Cells were incubated at 37°C for 3 h with shaking, and then isopropylthio--galactoside was added to a
final concentration of 0.5 mM. After growing overnight at
room temperature, the induced cells were pelleted by centrifugation and
were then resuspended in buffer containing 50 mM Tris-HCl
(pH 8.0), 150 mM NaCl, and 0.25 mM
phenylmethylsulfonyl fluoride and were then pelleted by centrifugation.
The cells were lysed by sonication in the resuspension buffer, and cell
debris was removed by centrifugation at 12,000g for 5 min. Proteins in the supernatant were assayed for PLD activity.
PLD Activity Assays
The basic PLD assay mixture contained 100 mM
Tris-HCl (pH 7.0), 80 mM KCl, 100 µM
Ca2+, 0.4 mM lipid vesicles, and 30 µg of
pBluescript SK-expressed protein in a total volume of 100 µL. Lipid
vesicles were composed of 35 nM of PE, 3 nM
PIP2, and 2 nM PC. PLD-mediated hydrolysis of
PC was measured using
dipalmitoylglycero-3-phospho-[methyl-3H]choline as
substrate (Pappan et al., 1997a). The reaction was initiated by adding
enzyme proteins and was incubated at 30°C for 30 min in a shaking
water bath. The reactions were stopped by adding 1 mL of
chloroform:methanol (2:1) and then 100 µL of 2 M KCl, and
the release of the [3H]choline into the aqueous phase was
quantitated by scintillation counting. Control assays were performed
using 30 µg of protein from lysed bacteria harboring the pBluescript
SK(-) plasmid without a PLD cDNA insert. The background activity from
bacteria was negligible and was subtracted from the activity of the
samples containing the recombinant PLD. To test the dependence of
PLD1 activity on divalent cations, 2 mM EGTA and 2 mM EDTA were added to the reactions with no
CaCl2 or MgCl2. In addition, CaCl2
or MgCl2 was added to the reaction at indicated
concentrations (Fig. 7).
To test the substrate selectivity of PLD1, mixed vesicles composed
of PIP2 (6.5 mol %) and PC, PE, or PS (93.5 mol
%) were used as substrates. Hydrolysis of PC, PE, or PS was
measured by the release of [3H]choline,
[14C]ethanolamine, or [14C]Ser.
[3H] PC was included at 18 nCi/reaction, and
[14C]PE or [14C]PS was included at 9 nCi/assay. Divalent cations and chelators were not added to the
reactions.
Dipalmitoylglycero-3-P-[methyl-3H]choline,
dioleoyl-glycero-3-P-[ethanolamine-2-14C]ethanolamine,
and
dioleoyl-glycero-3-P-[Ser-3-14C]Ser were
obtained from Amersham Biosciences (Piscataway, NJ). PC, PE, and PS
were purchased from Avanti Polar Lipids (Birmingham, AL).
| |
FOOTNOTES |
|---|
Received October 10, 2001; returned for revision November 8, 2001; accepted December 4, 2001.
1 This work was supported by the National Science Foundation (grant no. IBN-9808729) and the U.S. Department of Agriculture (2001-35304-10087). This is contribution 02-142-J of the Kansas Agricultural Experiment Station.
* Corresponding author; e-mail wangs{at}ksu.edu; fax 785-532-7278.
Article, publication date, and citation information can be found at www.plantphysiol.org/cgi/doi/10.1104/pp.010928.
| |
LITERATURE CITED |
|---|
|
TOP
ABSTRACT
INTRODUCTION
RESULTS AND DISCUSSION
MATERIALS AND METHODS
LITERATURE CITED
|
|---|
, in dehydration-inducible accumulation of phosphatidic acid in stress signaling.
Plant J
26: 595-605[CrossRef][ISI][Medline], -, and -.
Arch Biochem Biophys
353: 131-140[CrossRef][ISI][Medline], from Arabidopsis.
J Biol Chem
272: 7055-7061 is an acidic phospholipase active at near-physiological Ca2+ concentrations.
Arch Biochem Biophys
368: 347-353[CrossRef][ISI][Medline] and regulation of plant PLD, -, and - by polyphosphoinositides and calcium.
J Biol Chem
272: 28267-28273 phospholipid-binding C2 domain complex with Ca2+.
Structure
6: 1395-1405[Medline] and .
J Biol Chem
275: 19700-19706This article has been cited by other articles:
|
|
 |
|
  Y. Hong, S. Zheng, and X. Wang Dual Functions of Phospholipase D{alpha}1 in Plant Response to Drought Mol Plant, March 1, 2008; 1(2): 262 - 269. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Hong, X. Pan, R. Welti, and X. Wang Phospholipase D{alpha}3 Is Involved in the Hyperosmotic Response in Arabidopsis PLANT CELL, March 1, 2008; 20(3): 803 - 816. [Abstract] [Full Text] [PDF]
|
|
|
|
|
1 with
Distinct Regulatory Domains
