(Received for publication, November 15, 1996, and in revised form, January 2, 1997)
From the Department of Biochemistry, Kansas State University, Manhattan, Kansas 66506
A novel plant phospholipase D (PLD; EC 3.1.4.4)
activity, which is dependent on phosphatidylinositol 4,5-bisphosphate
(PIP2) and nanomolar concentrations of calcium, has been
identified in Arabidopsis. This report describes the
cloning, expression, and characterization of an Arabidopsis
cDNA that encodes this PLD. We have designated names of PLD for
this PIP2-dependent PLD and PLD
for the
previously characterized PIP2-independent PLD that requires
millimolar Ca2+ for optimal activity. The PLD
cDNA
contains an open reading frame of 2904 nucleotides coding for a
968-amino acid protein of 108,575 daltons. Expression of this PLD
cDNA clone in Escherichia coli results in the
accumulation of a functional PLD having PLD
, but not PLD
,
activity. The activity of the expressed PLD
is dependent on
PIP2 and submicromolar amounts of Ca2+,
inhibited by neomycin, and stimulated by a soluble factor from plant
extracts. Sequence analysis reveals that PLD
is evolutionarily divergent from PLD
and that its N terminus contains a regulatory Ca2+-dependent phospholipid-binding (C2) domain
that is found in a number of signal transducing and membrane
trafficking proteins.
Phospholipase D (PLD; EC 3.1.4.4)1-catalyzed hydrolysis of glycerophospholipids produces phosphatidic acid (PA) and a hydrophilic constituent. This activity was first identified in plants and since has been found in animals and microorganisms. PLD in plants was originally proposed to be important in phospholipid catabolism, initiating a lipolytic cascade in membrane deterioration during senescence and stress injuries (1, 2). Recent studies in plants, animals, and yeast indicate that PLD hydrolysis plays a pivotal role in transmembrane signaling and cellular regulation (3-9). Activation of PLD has been proposed to mediate many cellular processes including cell proliferation, membrane trafficking, meiosis, and responses to external and internal stimuli. It has been suggested that multiple forms of PLD are involved in these diverse cellular processes since several studies have shown the presence of PLD variants that are expressed differently (9-12). In castor bean (9) and rice (12), one PLD variant is constitutive whereas the appearance of other variants is associated with specific conditions such as rapid growth, wounding, and senescence. A distinct property shared by these variants is their in vitro requirement of millimolar Ca2+ concentrations for optimal activity. Further analyses of the castor bean PLD variants have led to the suggestion that the catalytic activity of these variants results from the same gene product (9-11).
A recent study has provided important evidence for the presence of two
plant PLDs that are derived from different gene products and regulated
distinctly (13). One PLD requires polyphosphoinositides and
submicromolar concentrations of Ca2+ for activity and the
other is PIP2-independent and is most active in the
presence of millimolar amounts of Ca2+. The latter is the
prevalent form of PLD that has been purified and characterized from a
number of plant species (14). Its cDNA has been recently cloned
from castor bean (15), Arabidopsis (16), rice and maize
(17). We have genetically suppressed the expression of this prevalent
plant PLD by introducing a PLD antisense gene into
Arabidopsis (13). While they showed less than 3% of the
PIP2-independent, millimolar Ca2+-requiring PLD
activity of wild-type Arabidopsis, the transgenic plants had
PIP2-dependent PLD activity comparable to that
of wild type at submicromolar calcium. In the present study, we provide molecular evidence for the presence of two distinct PLDs by isolating a
new PLD cDNA encoding the PIP2-dependent
PLD. Furthermore, analysis of the sequence and expressed protein from
the PLD cDNA gives further insights to the activation and function
of PLD in plants. Because the regulatory and structural features of
the newly identified PLD are distinct from those of the conventional
PLD, we have given names of PLD
for this
PIP2-dependent PLD and PLD
for the
previously characterized PIP2-independent PLD that requires
millimolar Ca2+ for optimal activity.
PIP and PIP2 were obtained from Boehringer Mannheim. Phosphatidylethanol, PI, and PE were purchased from Avanti Polar Lipids. All other phospholipids were obtained from Sigma. 1-Palmitoyl-2-oleoyl-[oleoyl-1-14C]glycero-3-P-choline and dipalmitoylglycero-3-P-[methyl-3H]choline were from DuPont NEN. Silica Gel 60 TLC plates were obtained from Merck (Darmstadt, Germany).
PLD cDNA Cloning and SequencingPutative expression
sequence-tagged Arabidopsis PLD cDNAs were identified by
searching the BLAST data base against the castor bean PLD cDNA
sequence. These clones were kindly provided by the Ohio State
University Arabidopsis Information Center. Strategies for
isolating full-length PLD cDNAs are described in the text. The
reactions for PCR amplification used DNA purified from an Arabidopsis PRL2 cDNA library (18) as DNA template, T7
sequence primer as the 5 primer, and 28 nucleotide bases corresponding to the near 5
end of the EST cDNA sequence as the 3
primer. The
reaction mixture consisted of 50 pmol of each primer, 0.5 µg of
template DNA, 50 mM KCl, 10 mM Tris-HCl (pH
9.0), 0.1% Triton, 2.5 mM MgCl2, 0.2 mM of each dNTP, 1 unit of Taq DNA polymerase in
a 100-µl volume. The thermal cycling was performed after an initial
denaturing cycle of 5 min at 95 °C. Then 25 to 30 cycles were
completed using the following temperature profiles: denaturation at
95 °C for 1 min, annealing for 30 s at 2-5 °C lower than
the calculated primer Tm, and extension for 1 min at
72 °C. PCR products were cloned into the pGEM-T vector
(Promega) according to the manufacturer's instructions.
RACE for 5-cDNA ends was performed according to the
manufacturer's instructions (Life Technologies, Inc.). The first
strand cDNA was synthesized from total RNA isolated from
Arabidopsis flowers. After PCR amplification using nested
gene-specific primers at the 3
end and a 5
-RACE anchor primer, the
DNA products were cut with KpnI and PstI and were
ligated into pBluescript (SK). The KpnI site was engineered
into the 5
primer, and the PstI was an internal site of the
PLD cDNA near the 3
end of the RACE product. To isolate
full-length PLD cDNAs a ZapII cDNA library, constructed from 3 to 6 kb mRNA isolated from hypocotyls of 3-day-old Arabidopsis seedlings (19), was screened using the 5
cDNA fragment generated by the 5
-RACE procedure. The hybridization
was conducted at 65 °C, and the subsequent DNA manipulation of the
positive clones was based on the previously described procedures
(15).
To sequence PLD clones, cDNA inserts from positive clones were digested with various restriction enzymes and the fragments were subcloned into the pBluescript plasmids, SK and/or KS. The complete DNA sequence was determined by using the Sequenase 2 kit according to the manufacturer's instructions (U. S. Biochemical Corp.). Vector pBluescript-based primers, universal forward and reverse, T3, T7, SK, and KS primers were used in most sequencing reactions. PLD cDNA-based primers were also synthesized for clarifying ambiguities. The final sequence was determined from both strands. Phylogenic analyses, pI calculations, and comparison of PLD nucleic acid and amino acid sequences were done with the Genetics Computer Group software (University of Wisconsin).
Expression of PLD cDNA in Escherichia coliExpression
of the PLD cDNA was performed using pBluescript SK(
)
containing the cDNA insert in E. coli. The recombinant
plasmid was transformed into E. coli JM 109. Fifty
microliters of an overnight culture of the transformed E. coli were added to 25 ml of LB medium with 50 µg/ml ampicillin.
The cells were incubated at 37 °C with shaking for 3 h, and
then IPTG was added to a final concentration of 2 mM. After
growing overnight at 30 °C, the induced cells were pelleted by
centrifugation and then resuspended in buffer containing 50 mM Tris-HCl (pH 8.0), 150 mM NaCl, 0.25 mM phenylmethylsulfonyl fluoride, 2 mM EDTA and
then pelleted by centrifugation. The cells were lysed by sonication in
the resuspension buffer and cell debris was removed by centrifugation
at 10,000 × g for 5 min. Proteins in the supernatant
were assayed for PLD activity and also subjected to SDS-PAGE followed
by immunoblot analysis using anti-PLD antibodies.
A 13-amino
acid peptide was synthesized that consisted of a cysteine and 12 amino
acids corresponding to the C terminus of PLD. The cysteine was added
to the N terminus of the peptide to provide a sulfhyryl group that was
used for conjugation of the peptide to keyhole limpet hemocyanin by
m-maleimidobenzoyl N-hydroxysuccinimide ester
(20). The conjugated protein was used as an immunogen to raise
antibodies in rabbits. For immunoblot analysis, proteins were separated
by 8% SDS-PAGE gels, transferred onto a polyvinylidene difluoride
membrane, and incubated with antiserum that contained JM109 lysate to
remove nonspecific reactive bacterial proteins. The immunoblot analysis
was performed as described (21).
PLD activity was assayed by using either 1-palmitoyl-2-oleoyl-[oleoyl-1-14C]glycero-3-P-choline or dipalmitoylglycerol-3-P-[methyl-3H]choline as substrates. The acyl-labeled PC was used for assaying transphosphatidylation activity whereas the choline-labeled PC was used in all other studies. In both cases, 2.5 µCi of radiolabeled PC was mixed with 3.5 µmol of PE, 0.3 µmol of PIP2, and 0.2 µmol of unlabeled PC in chloroform, and the solvent was evaporated under a stream of N2. In the phospholipid-specificity experiments, PIP2 was replaced with 0.3 µmol of PE, PA, PG, PS, PI, or PIP. The phospholipid substrate was dispersed in 1 ml of H2O by sonication at room temperature. Previously reported conditions were adapted to yield an enzyme assay mixture that contained 100 mM MES (pH 7.0), 5 µM CaCl2, 2 mM MgCl2, 80 mM KCl, 0.4 mM lipid vesicles, and 5-15 µg of expressed protein in a total volume of 100 µl (22). In the Ca2+ dependence experiments, the concentrations of free Ca2+ and Mg2+ in the reaction mixture were determined using Ca2+/Mg2+-EGTA buffers at pH 7.5 as described (23). The reaction was initiated by addition of substrate and incubated at 30 °C for 30 min in a shaking water bath. When choline-labeled PC was used, the reaction was stopped by addition of 1 ml of 2:1 (v/v) chloroform:methanol and 100 µl of 2 M KCl. After vortexing and centrifugation at 12,000 × g for 5 min, a 200-µl aliquot of the aqueous phase was mixed with 3 ml of scintillation fluid, and the release of [3H]choline was measured by scintillation counting.
When acyl-labeled PC was used, the reaction mixture included ethanol to a final concentration of 0.5% (v/v) for assaying the transphosphatidylation activity of PLD. The reaction was stopped by adding 375 µl of 1:2 (v/v) chloroform:methanol. Additionally, 100 µl of chloroform and 100 µl of 2 M KCl were added and the sample was vortexed. The chloroform and aqueous phases were separated by centrifugation at 12,000 × g for 5 min. The aqueous phase was removed and the chloroform phase was dried. Thin layer chromatography was conducted as described previously using 65:35:5 chloroform:methanol:NH4OH as the developing solvent (21). Lipids separated on plates were visualized by exposure to iodine vapor. Spots corresponding to lipid standards, PA, PC, and phosphatidylethanol, were scraped and radioactivity was measured by scintillation counting.
High Ca2+-dependent PLD Activity AssayThis assay reaction mixture contained 100 mM MES (pH 6.5), 25 mM CaCl2, 0.5 mM SDS, 1% (v/v) ethanol, 5-15 µg of protein, and 2 mM PC (egg yolk) containing dipalmitoylglycero-3-P-[methyl-3H]choline. The substrate preparation, reaction conditions, and product separation were based on previously described procedures (21) with the following changes: the assay volume was reduced to 100 µl and 100 µl of 2 M KCl was added to the 2:1 (v/v) chloroform:methanol extraction. The release of [3H]choline into the aqueous phase was quantitated by scintillation counting.
Arabidopsis Cytosolic and Membrane Fractionation; Heat and Protease TreatmentsLeaves from PLD antisense-suppressed
Arabidopsis (4 weeks old) were homogenized in a buffer
containing 50 mM Tris-HCl (pH 8.0), 10 mM KCl,
1 mM EDTA, 0.5 mM phenylmethylsulfonyl
fluoride, and 2 mM dithiothreitol. The cytosolic protein
fraction was the supernatant obtained after centrifugation of the
homogenate at 100,000 × g for 1 h. The pellet was
extracted with 0.44 M KCl in the homogenization buffer to
obtain the salt-solubilized membrane proteins (13). In typical PLD
assays, the cytosol containing 3 µg of protein was added to 10 µg
of the bacterially expressed PLD
. In the heat denaturation
treatment, the soluble fraction was boiled for 5 min followed by
centrifugation for 5 min at 12,000 × g to remove
precipitates. The clarified fraction was used directly or treated with
different proteases, thermolysin, trypsin, or proteinase K, to digest
proteins. After incubating at 37 °C for 30 min, trypsin and
proteinase K were inactivated by adding phenylmethylsulfonyl fluoride,
and thermolysin was inactivated by adding 2 mM EDTA.
Total RNA and genomic DNA
were isolated from Arabidopsis tissues (15). Full-length
cDNAs of PLD and PLD
were used as probes to hybridize the
genomic DNA digested with various restriction enzymes at 65 °C under
the previously described conditions (15). Total RNA was separated by
denaturing formaldehyde-agarose gel electrophoresis, transferred onto a
nylon membrane, and hybridized with a full-length PLD
cDNA at
65 °C (15).
Arabidopsis
expression sequence-tagged cDNA clones were identified as putative
PLD cDNAs by searching the BLAST data base using the castor bean
PLD cDNA sequence. These clones were about 1 kb in length and
incomplete PLD cDNAs. The first complete Arabidopsis PLD
cDNA was cloned by PCR using nested primers and encodes a protein
of 809 amino acids (16). This cDNA has been shown to be PLD,
because introduction of this cDNA as an antisense gene almost
completely abolishes the PLD
activity of transgenic plants. The
antisense plants lost the millimolar Ca2+-responsive PLD
activity, but showed PIP2-dependent PLD
activity comparable to that of wild-type plants at nanomolar
Ca2+ concentrations (13). In addition, the amino acid
sequence of the Arabidopsis PLD cDNA shares a high level
of identity (about 80%) with the previously cloned castor bean PLD
cDNA whose product displays PIP2-independent activity.
The cloning of PLD cDNA proved to be much more difficult than
that of PLD
. The cloning strategies involved nested PCR, 5
-RACE, and screening of cDNA libraries. A 1.5-kb 5
-fragment of the
cDNA was first cloned by PCR amplification using an
Arabidopsis PRL2 cDNA library (18) as a DNA template, an
internal fragment of the EST cDNA as the 3
primer, and T7 sequence
primer as the 5
primer. Further attempts at using nested PCR were not
fruitful, so 5
-RACE was performed to generate the missing 5
end. The
full-length cDNA was finally isolated by using the 5
-fragments to
screen a ZapII Arabidopsis cDNA library constructed
using 3-6 kb mRNA isolated from hypocotyls of 3-day-old
Arabidopsis seedlings (19).
The newly cloned PLD cDNA consists of 3309 nucleotides with an open
reading frame from 178 to 3081. The deduced amino acid sequence of this
protein is about 40% identical and 60% similar to that of
Arabidopsis PLD (Fig. 1) as well as to
PLDs cloned from castor bean, maize, and rice (15-17). The sequence
identity of this new cDNA is in contrast to previously cloned PLD
cDNAs from Arabidopsis, castor bean, maize, and rice,
which share about 75-90% amino acid sequence identity. All of these
previously cloned PLDs display PIP2 independent activity at
millimolar Ca2+ concentrations and thus belong to the class
of PLD
. Additionally, the newly cloned PLD differs from PLD
in
size and pI. PLD
s from different plant species are very similar in
size, ranging from 92 to 96 kDa, whereas the predicted polypeptide from
this cDNA has a calculated molecular mass of 109 kDa. This PLD
protein consists of 968 amino acids, which is 148 amino acids longer at
the N terminus than that of Arabidopsis PLD
(Fig. 1). The
newly cloned PLD has a calculated pI of 7.9 whereas all previously
cloned plant PLDs display acidic pI values (15-17). Prior to this
report, only PLDs from non-plant sources had been reported to have
basic pI values (4, 24). This newly cloned cDNA contains a
duplicated HXKXXXXD motif (amino acids 484-491
and 818-826, Fig. 1) that has been conserved in all PLDs and is
proposed to be involved in catalysis (24, 25). These results indicate
that the newly cloned PLD is a distinct isoform.
Establishing That the cDNA Encodes a New PIP2-dependent PLD
To establish
unequivocally that the newly cloned cDNA encodes a PLD, protein
from this cDNA was expressed in E. coli using pBluescript SK() as expression vector. After IPTG induction, the
production of a protein encoded by the cDNA was detected by immunoblotting using antibodies raised against a synthetic peptide corresponding to the 12 C-terminal amino acids of this protein (Fig.
2). No immunoreactive proteins were detected in the
protein extracts from E. coli containing vector alone, and a
trace amount of PLD was expressed without IPTG induction in the SK
construct.
The expressed protein was assayed for both PIP2-dependent and PIP2-independent PLD activity. There were only trace levels of PLD activities in protein extracts from E. coli JM109 harboring the SK alone or the vector containing the cDNA insert without IPTG induction. A significant increase in PIP2-dependent PLD activity was observed after IPTG induction (Fig. 2). The levels of PLD activity were in agreement with the presence or absence of PLD protein detected by immunoblotting. On the other hand, the expressed PLD showed no PIP2-independent, conventional PLD activity.
The transphosphatidylation activity of PLD was examined in order to
verify that the activity monitored in the bacterial extracts was due to
the action of PLD. IPTG-induced samples assayed in the presence of
0.5% ethanol showed a 13-fold increase in phosphatidylethanol production compared to uninduced samples (data not shown). Similarly, phosphatidic acid production increased 11-fold in IPTG-induced samples
relative to the uninduced samples, showing that the expressed protein
is indeed a member of the PLD family.
Recent studies have
shown that some mammalian PLDs are stimulated by cytosolic factors (22,
24). To examine whether PLD could be activated by plant soluble
factors, the expressed PLD
was assayed in the presence or absence of
a soluble fraction from Arabidopsis. The soluble extract was
obtained from transgenic Arabidopsis plants in which the
expression of PLD
was antisense suppressed. These plants were used
because the 100,000 × g supernatant contained
virtually no detectable PIP2-dependent nor
PIP2-independent PLD activity (13). The soluble extract
alone showed little PLD
activity (Fig.
3A), but its inclusion increased the
PIP2 dependent activity of the expressed protein and this
enhancement was dependent upon the concentration of cytosol added (Fig.
3B). These results suggest that the
PIP2-dependent PLD expressed in E. coli is stimulated by a soluble factor. In comparison, the
cytosolic fraction had no stimulatory effect on PLD
expressed from
its cDNA in E. coli (data not shown). Similarly, the
PIP2-dependent PLD extracted from the PLD
antisense membranes was insensitive to the addition of cytosol (Fig.
3B). It is possible that the cytosolic stimulator remains
bound with the membrane-associated PLD from the antisense plants.
The cytosolic factor was examined by various means to determine its
nature. The cloned human PLD is stimulated by small GTP-binding proteins of the ADP-ribosylation factor (ARF) family (24). We tested
the ability of recombinant human ARF1 to substitute for the cytosolic
factor from Arabidopsis. The amino acid sequences of human
ARF1 and Arabidopsis ARF are 88% identical and 95% similar (26). While it stimulated the expressed human PLD greatly, the recombinant human ARF1 had no effect on the expressed plant PLD (data not shown). The inclusion of GTP
S or GDP
S, which are
non-hydrolyzable analogues of GTP and GDP, respectively, in the
activity assays neither stimulated nor inhibited the level of
PIP2-dependent PLD activity (Fig.
3A). To examine further the nature of the cytosolic stimulatory factor, proteins in the soluble fraction were
heat-denatured. The stimulation of the expressed PLD by the cytosol
remained unchanged after the soluble fraction was boiled for 5 min
(Fig. 3A). Furthermore, treatment of the heat-denatured
cytosol with the proteases thermolysin, trypsin, and proteinase K also
did not inactivate the stimulatory effect of the cytosol (data not
shown). However, when the cytosol was passed over a gel filtration
column with a molecular mass cut-off of 6,000 Da, it lost its ability
to stimulate PLD
activity (Fig. 3A). Taken together,
these results suggest that the stimulating factor is a heat-stable
small molecule. The exact nature of the cytosolic factor remains to be
determined.
The requirement of specific phospholipids for PLD activity
was examined by substituting PIP2 in the lipid substrate
vesicles with PIP, PI, PG, PS, PE, or PA. PIP also was capable of
stimulating PLD
activity but its stimulation was only a third of the
level observed in the presence of PIP2 (Fig.
4A). The maximal stimulation of PLD
activity by PIP2 was achieved when PIP2 was
included in the substrate vesicles at an amount of 7.6 mol % (Fig.
4B). PIP2 levels much higher or lower than this
resulted in a significant decrease in activity. Further evidence for
the requirement of PIP2 for this PLD was obtained by
studying the influence of neomycin on its activity. Neomycin is a high
affinity cationic ligand for polyphosphoinositides and has been used to
demonstrate the activity dependence of various enzymes by it ability to
sequester PIP2 (27, 28). PLD
was inhibited by neomycin
in a concentration-dependent manner (Fig. 4C). PLD
activity was inhibited by greater than 50% at 500 µM
neomycin and was nearly abolished at 2 mM neomycin.
Catalytic properties of PLD expressed in
E. coli. A, effects of phosphoinositides and
other phospholipids on PIP2-stimulated PLD activity. Lipid
vesicles (0.4 mM) in the PIP2 requiring PLD assays were composed of 87 mol % PE, 7.6 mol % PIP2, and
5.4 mol % PC. PIP2 was replaced with 7.6 mol % PE, PA,
PG, PS, PI, or PIP. B, PIP2-stimulated PLD
activity as a function of mole % PIP2 in lipid vesicle.
Lipid vesicles (0.4 mM) in the reaction mixture consisted
of 79.4-94.6 mol % PE, 0-15.2 mol % PIP2, and 5.4 mol % PC. C, neomycin inhibition of PLD activity. PLD was assayed in the presence of 7.6 mol % PIP2 and 0-2 mM neomycin. D,
dependence of PLD
activity on Ca2+. Free
Ca2+ and Mg2+ in reaction mixtures were
controlled using Ca2+/Mg2+-EGTA buffers at pH
7.5. PLD
was expressed from its cDNA in E. coli, and
each assay used 10 µg of soluble protein containing PLD
and 3 µg
of cytosol proteins from PLD
antisense Arabidopsis. Values are means ± S.E. of three experiments.
To determine the influence of Ca2+ on
PIP2-dependent PLD activity, free
Ca2+ and Mg2+ in the reaction mixture were
controlled using Ca2+/Mg2+-EGTA buffers at pH
7.5 (23). PIP2 dependent activity was undetectable in the
absence of Ca2+, with little activity observed at or below
a concentration of 50 nM (Fig. 4D). At 500 nM calcium, PLD activity increased to a maximum and
gradually tapered off as millimolar levels of calcium were approached.
Under the optimal PIP2 and Ca2+ conditions, the
expressed PIP2-dependent PLD showed the highest activity between pH 7.0 and 7.5.
PLD from castor bean was reported to contain a C2 domain
at its N terminus (29). The C2 domain is a
Ca2+/phospholipid-binding domain present in a number of
different proteins involved in signal transduction and membrane
trafficking (29, 30). The three-dimensional structure of a C2 domain
from the neuronal protein synaptotagmin has been resolved recently by
x-ray crystallography and NMR (30, 31). The crystal structure of a
phosphoinositide-specific phospholipase C, which also contains a C2
domain, has been reported recently (32). The C2 domains of
synaptotagmin and PLC are comprised of an eight-strand sandwich containing 4-5 acidic residues involved in Ca2+ binding.
While the eight strands are conserved, the PLD
s from castor bean and
Arabidopsis possess substitutions within the C2 Ca2+-binding site, indicating a potential loss of
Ca2+ affinity (29) (Fig. 5).
An approach similar to that described previously (29) was used to align
the sequence of PLD with those of synaptotagmin and PLD
. The
PIP2-dependent PLD
contains a C2 domain near
its N terminus stretching from amino acid 158 to 279 (Fig. 5). The two
most highly conserved segments in different C2-containing proteins are
PYV and NPVFNEXF (30). These two regions have been proposed
to maintain the structural integrity of the C2 fold because their
residues are largely hydrophobic. In PLD
the first segment is
completely conserved and the second segment, NPVWMQHF, is largely the
same with several conservative amino acid substitutions. Furthermore, PLD
contains the conserved acidic residues (underlined in
Fig. 5) that serve to coordinate Ca2+-binding in the C2
domain. This is in contrast to PLD
in which two of the acidic
residues are substituted with positively charged or neutral amino
acids.
The molecular
organization of PLD and PLD
in the Arabidopsis genome
was examined by Southern blotting analysis (Fig. 6, A and B). Total genomic DNA was digested with
restriction enzymes and hybridized with probes made from full-length
PLD
and PLD
cDNAs. Hybridization of the same DNA with PLD
or PLD
probes gave unique banding patterns, indicating that the
PLD
and PLD
sequences do not cross-hybridize with each other at
high stringency conditions. The PLD
cDNA has one XhoI
site, but no BamHI, KpnI, and XbaI
restriction sites, and the digested genomic DNA gave one strong
hybridization band (Fig. 6A). The simple banding pattern of
hybridization by the PLD
cDNA indicates that the
Arabidopsis genome may contain one gene copy of PLD
. When
the same DNA was probed with PLD
, which contains one KpnI
and two BamHI recognition sites, but no XbaI nor
XhoI site, the number of hybridization bands were more than
that predicted from the BamHI, KpnI, and XhoI digestion of PLD
cDNA (Fig. 6B).
These results could be caused by the presence of these restriction
sites in intron sequences of the PLD
gene and/or by the presence of
another or closely related PLD
gene. Northern blot analysis using
the full-length PLD
cDNA as a probe detected one RNA band of
approximately 3.5 kb (Fig. 6C). The estimated size was in
agreement with that of the cloned PLD
cDNA. However, it was
unclear if this band was composed of one PLD
mRNA species or
different PLD transcripts with similar sizes. It is worth noting that
another study has shown that two protein bands from
Arabidopsis extracts were recognized by the antibodies
raised against a PLD
peptide, suggesting the presence of other
PLD(s) closely related to the cloned PLD
in Arabidopsis
(13).
The present results provide molecular evidence for the presence of
two plant PLD isoforms that are distinctly regulated and expressed.
PLD, which was previously cloned and characterized from several
plant species, requires millimolar Ca2+, but no
PIP2, for activity. Our study involving antisense
suppression of the PLD
gene in Arabidopsis has unmasked
the presence of a PIP2-regulated PLD in plants. In this
study, we have cloned and functionally expressed the
PIP2-dependent PLD which is designated PLD
.
The biochemical properties of the
PIP2-dependent PLD expressed from the PLD
cDNA are almost the same as those identified in Arabidopsis protein extracts (13). Specifically, the
PIP2 requirement by the PLD from the plant extracts and the
cDNA expressed in E. coli can be partially substituted
by PIP, but not by PI, PS, PG, PE, or PA. The optimal pH for the
PIP2-dependent PLD from both sources is around
7 to 7.5. The PLD obtained from both sources requires Ca2+
and is fully active at submicromolar ranges of Ca2+. These
similarities suggest that the cloned PLD is the isoform responsible for
the PIP2 dependent activity measured in the extracts of
Arabidopsis.
The only difference between the two enzymes appears in the
Ca2+ effect at higher concentrations; the activity from PLD
expressed in E. coli decreased at millimolar
Ca2+ whereas the PLD examined in the PLD-antisense
plants showed a sigmoidal response to the increase of Ca2+
concentration. However, the plant extract contains more than one type
of PLD, and PLD
is known to be most active at millimolar Ca2+. Therefore, the stable PLD activity observed in
millimolar Ca2+ could result from residual PLD
or from
any other PLD that is stimulated by millimolar Ca2+ and not
suppressed by PLD
antisense.
The ability of PLD to shift from an inactive state to a highly
active state over a narrow range of calcium concentration strongly
suggests that changing intracellular concentrations could be a major
form of regulation for the enzyme in vivo. This observation is particularly relevant considering that 100 nM and 1 µM are the respective resting and stimulated
intracellular calcium concentrations of plants and animals (33, 34). It
remains to be elucidated how Ca2+ is involved in the
PLD-mediated hydrolysis of phospholipids. The finding of a C2 domain
near the N-terminal region of several PLDs suggests that one of the
roles of Ca2+ may be to regulate the enzymes' binding to
phospholipids. The predominant feature of the C2 domain is its ability
to mediate Ca2+-dependent phospholipid binding.
The Ca2+/phospholipid-binding domain was first identified
in Ca2+-dependent protein kinase C isoforms and
has since been found in a number of different proteins including
intracellular PLA2 and PIP2-PLC isoforms. It is
believed that the binding of membrane phospholipids by a C2 domain
represents a Ca2+-dependent translocation
mechanism whereby cytosolic proteins become associated with the
membrane in a highly regulated manner. Thus for some C2-containing
enzymes, phospholipid binding could represent one mechanism of cellular
activation. The presence of a C2 domain in PLDs raises the question of
whether Ca2+-dependent phospholipid binding is
involved in the enzyme's activation, catalysis, or both. Comparison of
the C2 domains of PLD
and PLD
reveals an important difference.
The PLD
C2 domain conserves the acidic residues needed to coordinate
Ca2+ binding whereas the PLD
C2 domain possesses
substitutions, potentially indicating a loss of Ca2+
affinity. The difference in the amount of Ca2+ needed for
activity is one of the most distinct in vitro properties that distinguishes between PLD
and PLD
. PLD
requires
millimolar amounts of Ca2+ whereas PLD
is fully active
at low micromolar levels of Ca2+ (13). An ongoing study in
this laboratory is to determine whether or not the differences in the
C2 domain underlie the different Ca2+ requirements observed
for PLD
and PLD
. Such studies should help understand the
regulatory and catalytic mechanisms for these PLD isoforms.
Sequence analysis indicates that PLD and PLD
are evolutionarily
divergent and that PLD
is more closely related to the PLDs cloned
from yeast (4) and human (24) than is PLD
. Alignments of these PLD
sequences reveals two distinct groups: PLDs from plants and those from
human and yeast. Within the plant group PLD
forms a subgroup
distinct from that of PLD
s from Arabidopsis, castor bean,
maize, and rice. The grouping of PLDs from Arabidopsis, castor bean, maize, and rice suggests that these are more closely related evolutionarily. Phylogram analysis in the unrooted phylogeny places the Arabidopsis PLD
with the yeast and human PLDs.
Furthermore, the phylogenic groupings are supported by comparing the
calculated pI values and catalytic properties of these PLDs. PLD
of
different plant species all have acidic pI values around 5-6 whereas
PLD
and the yeast and human PLDs have basic pI values of 7.9, 7.6, and 9.3, respectively. PLD
activity is PIP2-independent
and requires millimolar concentrations of Ca2+ whereas
PLD
and the cloned PLDs from human and yeast are activated by
PIP2. Both PLD
and the human
PIP2-dependent PLD are regulated by
physiological concentrations of Ca2+ and cytosolic
factors.
These analyses have clearly shown that there are PLD isoforms in plants
that are encoded by different genes and are regulated in a distinct
manner. The distinct regulatory mechanisms suggest that these PLDs have
different cellular functions. PLD is the most prevalent plant PLD
(13) and it seems to be unique to plants based on the comparison of its
catalytic properties with those of mammalian and yeast PLDs. PLD
, on
the other hand, shares some properties with the recently cloned human
and yeast PLDs. It has been shown that the yeast PLD is required for
meiosis (4-6). The cloned human PLD is thought to be involved in
membrane trafficking and secretion but its role in these processes is
unclear (3, 24). We have produced PLD
-suppressed transgenic plants
that should be instrumental in defining the function of PLD
. Efforts are underway to produce plants deficient in PLD
activity and to use
these systems to sort out the roles of the different PLDs in growth and
development.
The nucleotide sequence(s) reported in this paper has been submitted to the GenBankTM/EMBL Data Bank with accession number(s) U36381[GenBank] and U84568[GenBank].
We thank Dr. A. J. Morris for providing us the recombinant human PLD and ADP-ribosylation factor.